Global Metabolite Profiling of Human Colorectal Cancer Xenografts

(1-3) The objective in global metabolite profiling, also known variously as metabolomics or metabonomics,(4) is to unravel multiple complex molecular ...
23 downloads 12 Views 1MB Size
Article pubs.acs.org/jpr

Global Metabolite Profiling of Human Colorectal Cancer Xenografts in Mice Using HPLC−MS/MS Neil J. Loftus,*,† Lindsay Lai,‡,∥ Robert W. Wilkinson,§ Rajesh Odedra,§ Ian D. Wilson,#,‡ and Alan J. Barnes† †

Mass Spectrometry Business Unit, Shimadzu, Manchester, United Kingdom Department of Drug Metabolism and Pharmacokinetics IM, AstraZeneca, Mereside, Alderley Park, Macclesfield, Cheshire SK10 4TG, United Kingdom § Cancer and Infection Research Area, AstraZeneca, Mereside, Alderley Park, Macclesfield, SK10 4TG, United Kingdom ∥ Manchester Institute of Biotechnology, 131 Princess Street, Manchester, M1 7DN, United Kingdom ‡

S Supporting Information *

ABSTRACT: Reversed-phase gradient LC−MS was used to perform untargeted metabonomic analysis on extracts of human colorectal cancer (CRC) cell lines (COLO 205, HT-29, HCT 116 and SW620) subcutaneously implanted into age-matched athymic nude male mice to study small molecule metabolic profiles and examine possible correlations with human cancer biopsies. Following high mass accuracy data analysis using MS and MS/MS, metabolites were identified by searching against major metabolite databases including METLIN, MASSBANK, The Human Metabolome Database, PubChem, Biospider, LipidMaps and KEGG. HT-29 and COLO 205 tumor xenografts showed a distribution of metabolites that differed from SW620 and HCT 116 xenografts (predominantly on the basis of relative differences in the amounts of amino acids and lipids detected). This finding is consistent with NMR-based analysis of human colorectal tissue, where the metabolite profiles of HT-29 tumors exhibit the greatest similarity to human rectal cancer tissue with respect to changes in the relative amounts of lipids and cholinecontaining compounds. As the metabolic signatures of cancer cells result from oncogene-directed metabolic reprogramming, the HT-29 xenografts in mice may prove to be a useful model to further study the tumor microenvironment and cancer biology. KEYWORDS: metabonomics, metabolomics, tumor xenografts, colorectal cancer, LC−MS

Received: March 24, 2013

© XXXX American Chemical Society

A

dx.doi.org/10.1021/pr400260h | J. Proteome Res. XXXX, XXX, XXX−XXX

Journal of Proteome Research

Article



tumor progression under conditions of oxidative stress. Recent evidence also suggests that metabolites can be viewed as oncogenic mediators by affecting cell signaling and blocking cell differentiation.8 This has resulted in a new approach in understanding the role of the tumor microenvironment and developing new strategies for diagnostics and therapeutics to target and prevent cancer metabolism.9 The development of novel cancer therapeutics relies heavily on the use of model systems, in particular the use of mice bearing human tumor xenografts, to assess the clinical efficacy on human cancers in vivo.10,11 Helping to characterize tissue at the molecular level, by measuring differences and similarities between the metabolic profiles of human colorectal cancer (CRC) xenografts versus CRC patient biopsies, may prove to be an important step in determining the predictive value of human tumor xenografts, by enabling the selection of those that most resemble human tumor samples for selection as appropriate model systems in preclinical biomedical research. Metabolic profiling studies mainly involve the multicomponent analysis of biological samples using several techniques such NMR spectroscopy12 and/or mass spectrometry (LC−MS, GC− MS, CE−MS or MALDI).13 In this study, high mass accuracy LC− MS/MS was used to identify potential marker metabolites in human tumor samples using authentic standards, or where this was not possible, analytes were provisionally identified by searching for

INTRODUCTION A more global, systems-based approach to study changes in smallmolecule metabolite profiles and fluxes in biological matrices has become an important tool to understand the disease process and possible therapeutic treatments. This approach is considered to complement genomics, transcriptomics and proteomics.1−3 The objective in global metabolite profiling, also known variously as metabolomics or metabonomics,4 is to unravel multiple complex molecular networks that may be used to characterize the development and progression of a disease and identify appropriate biomarkers. In cancer, oncogenic events reorganize metabolic pathways so that tumor cells can increase nutrient uptake promoting biosynthetic capabilities and change the behavior of cells to increase their chance of survival.5−7 Both changes in the tumor microenvironment and metabolic reprogramming enable the tumor cell to achieve a more aggressive phenotype, which may reflect Table 1. Tumor Xenograft Sample Group Summary colorectal cell line tumor growth period (days) post implantation number of tumor tissue samples

HT-29

HCT 116

SW620 COLO 205

7 and 14

33

35

13

10 (2 × 5)

9

10

17

Figure 1. PCA plot from chromatogram matrix of organic extracts analyzed in positive ion mode. Pooled QC (closed gray squares); CRC cell lines: HT-29 (closed blue diamonds), COLO 205 (open green squares), SW620 (open red circles), HCT 116 (closed black triangles). B

dx.doi.org/10.1021/pr400260h | J. Proteome Res. XXXX, XXX, XXX−XXX

Journal of Proteome Research

Article

Animal (Scientific Procedures) Act 1986. Human tumor xenograft studies were conducted on male Swiss athymic nude mice (nu/nu genotype; AstraZeneca) between the ages of 8−10 weeks. Tumor xenografts were established by injecting tumor cell suspensions subcutaneously on the dorsal flank of mice, as previously described.22,23 Mice were humanely euthanized, and tumors were excised; Table 1 summarizes the sampling time postimplantation and sample numbers.

metabolites using MS and MS/MS databases such as METLIN [http://metlin.scripps.edu],14 MassBank [http://www.massbank. jp],15 The Human Metabolome Database [www.hmdb.ca/],16 PubChem [http://pubchem.ncbi.nlm.nih.gov/],17 Biospider [http://biospider.ca/],18 LipidMaps [www.lipidmaps.org],19 and KEGG [http://www.genome.jp/kegg/].20 The intent of this study was to determine the metabolic signatures of well-defined xenograft models of human CRC cell lines (COLO 205, HT-29, HCT116 and SW620) that best matched published human CRC tissue biopsy data using high mass accuracy LC−MS/MS data. The cell lines selected for this are among the most commonly used models representing this cancer type.21



Sample Preparation

Tumor samples were prepared via a two-step solvent extraction in which polar hydrophilic substances were separated into an aqueous phase, and nonpolar lipophilic compounds separated into an organic phase. For this 100 mg of manually disaggregated tumor tissue was mixed with 1 mL of acetonitrile−water (1:1 v/v) in an Eppendorf vial and sonicated for 5 min followed by centrifugation at 17900g for 10 min. The acetonitrile−water supernatants were then taken as aqueous extracts and stored in fresh Eppendorf vials at −20 °C until analyzed. The pellets were resuspended in 1 mL of chloroform−methanol (3:1 v/v) and sonicated for 5 min followed by centrifugation at 17900g for 10 min. The supernatants were taken into fresh Eppendorf vials and were left uncovered in a fume cupboard overnight to allow the evaporation of the organic solvents. These dry residues from the “organic extracts” were resuspended in 1 mL of methanol (4 °C)

EXPERIMENTAL SECTION

Animal Studies

Human CRC cell lines COLO 205, HCT 116, HT-29, and SW620 were obtained from the American Type Culture Collection (ATCC) (http://www.lgcstandards-atcc.org/ATCCCulturesandProducts/ CellBiology/CellLinesandHybridomas/tabid/981/Default.aspx) and cultured using standard media and conditions (http://www.atcc. org/CulturesandProducts/CellBiology/CellLinesandHybridomas/ tabid/169/Default.aspx). All animal experiments were conducted in full accordance with the United Kingdom Home Office

Figure 2. PCA plot from chromatogram matrix of aqueous extracts analyzed in negative ion mode. Pooled QC (closed gray squares); CRC cell lines: HT-29 (closed blue diamonds), COLO 205 (open green squares), SW620 (open red circles), HCT 116 (closed black triangles). C

dx.doi.org/10.1021/pr400260h | J. Proteome Res. XXXX, XXX, XXX−XXX

Journal of Proteome Research

Article

and aliquoted into vials. In addition 10 μL of each sample from the aqueous and organic extracts were taken and separately pooled to provide a bulk quality control (QC) sample for each type of extract as described previously.24,25

separate chromatographic runs, with the sample sequence randomized and a QC sample acquired every 5 samples. All MS, MS/MS mass spectra were acquired using the following method parameters: mass range, m/z 150−900 in MS and m/z 150−900 in MS/MS mode; ion source temperature, 230 °C; heated capillary temperature, 230 °C; ESI voltage, 4.5 kV; ESI nebulization gas flow, 1.5 L/min; detector voltage, 1.75 kV; ion accumulation time, 25 ms. Automated data-dependent functions were set to acquire five scans for each precursor detected using the most intense ion signal as the trigger. Positive and negative ion modes were used simultaneously with a polarity switching time of 100 ms. Mass calibration was carried out using a trifluoroacetic acid sodium solution (2.5 mmol L−1) from 100 to 1250 Da.

LC−MS/MS Analysis

LC−MS/MS analysis was performed using a Prominence LC system (Shimadzu Corporation, Kyoto, Japan) connected to a quadrupole ion trap time-of-flight mass spectrometer (LCMSIT-TOF, Shimadzu, Kyoto, Japan) equipped with an electrospray ionization (ESI) source. Extracted tumor samples were kept at 4 °C in the autosampler, and 2 μL was taken for analysis by reversed-phase gradient LC fitted with a Waters Symmetry 3.5 μm C18 column (150 × 2.1 mm). The chromatographic system used a binary solvent system delivered as a gradient of 0.1% (v/v) formic acid in water (solvent A) and 0.1% (v/v) formic acid in acetonitrile (solvent B) using a flow rate of 0.3 mL min−1 with the column maintained at 40 °C. The starting gradient conditions were 95% A:5% B with a linear gradient up to 95% B over 30 min. The solvent composition was then held at 95% B for 5 min, after which the column was returned to 5% B over the next 5 min, making a total cycle time of 40 min/sample. Column conditioning was performed by repeatedly injecting the QC sample at the beginning of the run prior to the sample run. In this study the QC sample was injected 10 times at the beginning of the run. Aqueous and organic samples were analyzed in

Data Analysis

To perform successful global comparisons, spectral and chromatographic alignment was performed using a nonlinear retention time alignment, filtration, peak detection and peak matching software tool (Profiler Solutions, Shimadzu, Kyoto, Japan). The resultant time and mass-aligned chromatographic data array was corrected to remove ion signals with poor response repeatability (a metabolite ion response variance of less than 20% RSD within the pooled biological QC sample was considered acceptable;26 the ion must also have been present in 80% of the QC data files). Data was then exported and analyzed by SIMCA-P (ver. 12.1) by PCA. Ions of interest were searched against external major

Table 2. Metabolites Identified by High Mass Accuracy MS/MS Analysis in Both Positive and Negative Ion Modea colorectal cell line compound

formula

Amino acids C10H18N4O6 N-(L-arginino)succinate Methionine C5H11NO2S Tyrosine C9H11NO3 Phenylalanine C9H11NO2 Tryptophan C11H12N2O2 Nucleoside Uridine C9H12N2O6 Sialylated tumor markers in humans N-Acetyl-a-neuraminic acid C11H19NO9 Coproduct of polyamine biosynthesis 5′-Deoxy-5′-(methylthio) C11H15N5O3S adenosine Vitamin Pantothenic Acid C9H17NO5 Choline contaning compounds Glycerophosphocholine | PC C8H20NO6P (0:0/0:0) Lipids Sphinganine C18H39NO2 Palmitoyl Ethanolamide C18H37NO2 O-Palmitoyl-R-carnitine C23H45NO4 Lyso-PE(0:0/20:5) C25H42NO7P PE(20:5(5Z,8Z,11Z,14Z,17Z)/ C25H44NO7P 0:0) PC(13:0/0:0) C21H44NO7P PC(16:0/0:0)[U]/PC C24H50NO7P (16:0/0:0)[rac] PC(18:1(9Z)/0:0) C26H52NO7P PE(18:1(9Z)/0:0) C23H46NO7P PC(16:0/0:0) C26H54NO7P

ion mode

m/z theoretical

tR (min)

(M + H)+ (M + H)+ (M + H)+ (M + H)+ (M + H)+

291.1299 150.0583 182.0812 166.0863 205.0972

1.06 1.30 1.46 2.47 3.32

(M − H)−

243.0623

(M + H)+

MS/MS data reference

HT29 avg pk area

COLO205 avg pk area

HCT116 avg pk area

SW620 avg pk area

MID 389 MID 26 MID 34 MID 28 MID 33

8 017 428 5 521 397 12 727 544 21 524 242 6 742 327

6 123 627 3 757 313 9 240 315 12 714 151 6 094 254

2 559 008 3 582 192 8 252 054 13 758 207 5 157 815

2 262 892 2 130 460 5 552 815 4 506 341 3 144 015

1.31

MB 90

59 927 767

76 184 638

20 464 590

29 333 545

310.1133

1.59

MID 5739

12 861 448

7 268 483

755 185

1 231 516

(M + H)+

298.0968

3.28

MID 3425

6 392 881

7 288 882

1 416 840

1 558 021

(M + H)+

220.1179

3.23

MID 241

5 263 455

10 248 678

797 607

2 717 985

(M + H)+

258.1101

1.16

MID 370

204 699 454

190 357 782

1 985 750

19 040 757

(M + H)+ (M + H)+ (M + H)+ (M − H)− (M − H)−

302.3050 300.2897 400.3421 498.2626 500.2928

14.30 15.24 17.61 18.99 20.42

MID 395 MID 43210 MID 961 UT001941 UT002199

8 833 084 29 608 819 52 950 639 5 140 303 17 687 484

8 942 481 51 032 090 27 898 006 3 497 938 17 689 216

9 067 950 48 399 381 107 779 658 1 429 583 12 962 403

9 364 102 27 956 828 61 068 291 1 369 416 14 635 855

(M + H)+ (M + H)+

454.2928 496.3398

22.09 22.20

LMGP01050001 MID 182

67 297 189 203 240 455

78 473 102 185 831 643

44 087 644 137 343 431

25 697 570 107 254 788

(M + H)+ (M + H)+ (M + H)+

522.3554 480.3085 524.3711

22.34 22.79 25.52

LMGP01050032 LMGP02050004 LMGP01050026

101 919 565 95 823 046 183 985 450

92 537 279 159 624 079 151 093 072

54 732 106 100 740 324 102 955 932

116 518 771 136 033 064 87 778 277

a

Metabolite identification was confirmed (where possible) using MS/MS fragment ion data by referencing major metabolite databases (for example, METLIN and MASSBANK). D

dx.doi.org/10.1021/pr400260h | J. Proteome Res. XXXX, XXX, XXX−XXX

Journal of Proteome Research

Article

metabolite databases including METLIN,14 MassBank,15 The Human Metabolome Database,16 PubChem,17 Biospider,18 LipidMaps19 and KEGG.20 For these ions, accurate mass information was used to calculate probable elemental composition using Formula Prediction software (Shimadzu, Kyoto, Japan) to analyze isotopic profiles of MS data.27 This information together with MS/MS collision induced dissociation (CID) data was used for metabolite identification.

each human cell line xenograft tumor tissue and to compare with published human CRC tumor metabolite data. To look for differences in the metabolite profiles found between the human CRC tissue xenografts and for subsequent comparison with published human CRC tumor metabolite data, a multivariate statistical strategy was used. As biological samples are inherently complex, matching peaks representing the same analyte from different samples requires data preprocessing approaches to integrate peak detection and peak alignment before further statistical analysis and verification. Although there are a number of data preprocessing tools available as open source and freely available software (XCMS,28 MathDAMP,29 Metalign,30 MZmine31), in this study Profiling Solution software was used to generate a crosssample peak-matched data array including sample and pooled quality control (QC) data. This software was developed to optimize data alignment in complex data sets using both positive and negative ion MS data and includes several tools to filter the data matrix processing parameters including deisotoping and applying QC criteria. Principal component analysis (PCA) showed separation between the organic and aqueous extracts of the COLO 205 and HT 29 CRC cell lines, both from each other and the HCT 116 and SW620 cell lines, which did not appear to be well differentiated from each other (in either ionization mode) as illustrated in Figures 1 and 2. Multivariate statistics, using tools such as the S-plot, helped to identify metabolite ion masses with major influence on the cell line group membership. HT-29 and COLO 205 extracts differed markedly from the other two human CRC cell lines in the relative



RESULTS AND DISCUSSION A number of mouse models of colorectal cancer (CRC) have been developed to investigate the biological mechanisms underlying the oncogenic process. Currently used models include a variety of genetically engineered, carcinogen-induced and xenograft mouse models, and it is generally agreed that no one model is sufficient to elucidate all aspects of CRC etiology. However, the most widely used in vivo model for evaluating new anticancer drug therapies is the xenograft model. Xenografts have been established through the subcutaneous injection of genetically defined human-derived cell lines into immune-compromised nude mice. The advantages of this model include a high degree of predictability and rapid tumor formation. Although there are a large number of xenograft cell lines used in CRC research, this study considered human cell lines COLO 205, HT-29, HCT116 and SW620, as they are among the most commonly used models representing this cancer type. The objective of the study was to determine the metabolite profiles in

Figure 3. The metabolic profiles detected by high accuracy LC−MS/MS analysis showed that HT-29 and COLO 205 differed markedly from HCT 116 and SW620 cell lines mostly through changes in amino acid, nucleoside and lipid metabolism (group average peak area response; the error bar represents +1 SD). E

dx.doi.org/10.1021/pr400260h | J. Proteome Res. XXXX, XXX, XXX−XXX

Journal of Proteome Research



Article

CONCLUSIONS The ability of metabolic profiling via LC−MS to readily differentiate between tumor xenografts derived from different humanderived CRC cell lines has been demonstrated. This tumor profiling methodology may therefore be of considerable value in studies designed to metabolically phenotype different tumors with a view to better understanding their biochemistry and may provide a useful method for molecular pathology. This approach may provide a new method for identifying appropriate animal models for certain human diseases and help understand the underlying pathology associated with these conditions. Should these differences translate into human disease, such information might provide a means for personalizing treatments based on the administration of drugs best suited to treating that particular phenotype.

amounts of several metabolites, most notably the up-regulation of amino acids (methionine, phenylalanine, tyrosine, tryptophan), fatty acids (arachidonic acid, palmitic acid, eicosapentaenoic acid), nucleotides (uridine) and lipids (lysophosphatidylcholines) (see Table 2 for relevant MS-identification data). The same changes in metabolite profiles were also reported in human CRC tissue biopsies32 and confirm previously published data using alternative analytical platforms (high-resolution magic angle spinning nuclear magnetic resonance (HR-MAS NMR) spectroscopy,33 GC-TOF34). Metabolite profiles in human CRC tissue are characterized by metabolic dysregulation in several metabolic pathways including glycolysis, nucleotide biosynthesis, lipid metabolism, inflammation and steroid metabolism, together with elevated tissue hypoxia.32 In the present study, the metabolic phenotypes of COLO 205, HCT 116, HT-29, and SW620 xenografts differed in several metabolic pathways; the most notable differences were observed in the relative amounts of N-acetylneuraminic acid (sialic acid), glycerophosphocholine and the distribution of amino acids (methionine, N-(L-arginino)succinate, phenylalanine, tyrosine and tryptophan). The clinical significance of high levels of N-acetylneuraminic acid has been widely reported in malignant human tumors, changing the expression of cell-surface glycans, especially an enrichment of certain sialic acid-containing antigens.35 The majority of sialylated tumor markers in humans involve changes in presentation of the common human sialic acid N-acetylneuraminic acid (Neu5Ac). Increased amounts of choline-containing compounds (ChoCC) such as phosphocholine, phosphatidylcholine and glycerophosphocholine, which are important precursor constituents of cell membranes, have been proposed as markers for cell proliferation and found to be elevated in a number of malignant lesions. The ethanolamine-containing components, ethanolamine, glycerophosphoethanolamine, and phosphoethanolamine are also major precursors and degradation products of membrane assembly and catabolism. HT-29 showed the highest amount of glycerophosphocholine; however, phosphoethanolamine was highest in the COLO 205 cell line. Amino acid (methionine, phenylalanine, tyrosine and tryptophan) up-regulation was also highest in the HT-29 extracts. Amino acid up-regulation has been previously reported in a quantitative metabolome profiling study of colon and stomach tumor tissue in cancer patients.36 In this study, concentrations of most amino acids and their primary metabolites were higher in tumor tissues compared to normal colon and stomach tissues.36 The differences highlighted between the various cell lines for these metabolites xenografts in the present study are illustrated in Figure 3. As indicated in the introduction, the intent of this study was to identify the metabolic signatures of well-defined xenograft models of human CRC that best matched published human CRC tissue biopsy data using high mass accuracy LC−MS/MS data. Our studies showed that the metabolite profiles from HT29 appeared to be most similar to human CRC tissues, although the metabolism of COLO 205 was also similar to HT-29. This finding is also in close agreement with NMR data, which enabled assignments of 27 metabolites in HT-29, HCT116 and SW-620 xenografts and considered HT-29 as the most appropriate human rectal xenograft model for preclinical studies.33



ASSOCIATED CONTENT

* Supporting Information S

Supplementary data for tables and figures. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +44 161 88 66 558. Fax: +44 161 88 66 559. Present Address #

Department of Surgery and Cancer, Imperial College, London SW7 2AZ, United Kingdom. Notes

The authors declare no competing financial interest.



REFERENCES

(1) Fernie, A. R.; Trethewey, R. N.; Krotzky, A. J.; Willmitzer, L. Metabolite profiling: from diagnostics to systems biology. Nat. Rev. Mol. Cell Biol. 2004, 5 (9), 763−769. (2) Nicholson, J. K.; Wilson, I. D. Understanding ‘global’ systems biology: metabonomics and the continuum of metabolism. Nat. Rev. Drug Discovery 2003, 2 (8), 668−676. (3) Griffin, J. L. The Cinderella story of metabolic profiling: does metabolomics get to go to the functional genomics ball? Philos. Trans. R. Soc., B 2006, 361 (1465), 147−161. (4) Nicholson, J. K.; Lindon, J. C. Systems biology: Metabonomics. Nature 2008, 455 (7216), 1054−1056. (5) Zaugg, K.; Yao, Y.; Reilly, P. T.; Kannan, K.; Kiarash, R.; Mason, J.; Huang, P.; Sawyer, S. K.; Fuerth, B.; Faubert, B.; Kalliomaki, T.; Elia, A.; Luo, X.; Nadeem, V.; Bungard, D.; Yalavarth, S.; Growney, J. D.; Wakeham, A.; Moolani, Y.; Silester, J.; Annick, T. Y.; Bakker, B.; Tsuchihara, K.; Berger, S. L.; Hill, R. P.; Jones, R. G.; Tsao, M.; Robinson, M. O.; Thompson, C. B.; Pan, G.; Mak, T. W. Carnitine palmitoyltransferase 1C promotes cell survival and tumor growth under conditions of metabolic stress. Genes Dev. 2011, 25, 1041−1051. (6) Stewart, J. D.; Marchan, R.; Lesjak, M. S.; Lambert, J.; Hergenroeder, R.; Ellis, J. K.; Lau, C. H.; Keun, H. C.; Schmitz, G.; Schiller, J.; Eibisch, M.; Hedberg, C.; Waldmann, H.; Ekkehart, L.; Tanner, B.; Sehouli, J.; Sagemueller, J.; Staude, H.; Steiner, E.; Hengster, J. G. Choline-releasing glycerophosphodiesterase EDI3 drives tumor cell migration and metastasis. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 8155−8160. (7) Mullen, A. R.; Wheaton, W. W.; Jin, E. S.; Chen, P. H.; Sullivan, L. B.; Cheng, T.; Yang, Y.; Linehaven, W. M.; Chandel, N. S.; DeBerardinis, R. J. Reductive carboxylation supports growth in tumor cells with defective mitochondria. Nature 2012, 481, 385−388. (8) Ward, P. S.; Thompson, C. B. Metabolic reprogramming: a cancer hallmark even warburg did not anticipate. Cancer Cell 2012, 21, 297− 308.

F

dx.doi.org/10.1021/pr400260h | J. Proteome Res. XXXX, XXX, XXX−XXX

Journal of Proteome Research

Article

metabonomic analysis: application to human urine. J. Proteome Res. 2007, 6 (8), 3291−3303. (26) Dunn, W. B.; Broadhurst, D.; Begley, P.; Zelena, E.; FrancisMcIntyre, S.; Anderson, N.; Brown, M.; Knowles, J. D.; Halsall, A.; Haselden, J. N.; Nicholls, A. W.; Wilson, I. D.; Kell, D. B.; Goodacre, R. Procedures for large-scale metabolic profiling of serum and plasma using gas chromatography and liquid chromatography coupled to mass spectrometry. Nat. Protoc. 2011, 6 (7), 1060−1083. (27) Loftus, N.; Miseki, K.; Iida, J.; Gika, H. G.; Theodoridis, G.; Wilson, I. D. Profiling and biomarker identification in plasma from different Zucker rat strains via high mass accuracy multistage mass spectrometric analysis using liquid chromatography/mass spectrometry with a quadrupole ion trap-time of flight mass spectrometer. Rapid Commun. Mass Spectrom. 2008, 22 (16), 2547−2554. (28) Smith, C. A.; Want, E. J.; O’Maille, G.; Abagyan, R.; Siuzdak, G. XCMS: Processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal. Chem. 2006, 78, 779−787. (29) Baran, R.; Kochi, H.; Saito, N.; Suematsu, M.; Soga, T.; Nishioka, T.; Robert, M.; Tomita, M. MathDAMP: a package for differential analysis of metabolite profiles. BMC Bioinf. 2006, 7, 530. (30) Lommen, A. MetAlign: Interface-driven, versatile metabolomics tool for hyphenated full-scan mass spectrometry data preprocessing. Anal. Chem. 2009, 81 (8), 3079−3086. (31) Katajamaa, M.; Miettinen, J.; Oresic, M. MZmine: toolbox for processing and visualization of mass spectrometry based molecular profile data. Bioinformatics 2006, 22 (5), 634−636. (32) Chan, E. C. Y.; Koh, P. K.; Mal, M.; Cheah, P. Y.; Eu, K. W.; Backshall, A.; Cavill, R.; Nicholson, J. K.; Keun, H. C. Metabolic profiling of human colorectal cancer using high-resolution magic angle spinning nuclear magnetic resonance (HR-MAS NMR) spectroscopy and gas chromatography mass spectrometry (GC/MS). J. Proteome Res. 2008, 8 (1), 352−361. (33) Seierstad, T.; Roe, K.; Sitter, B.; Halgunset, J.; Flatmark, K.; Ree, A.; Olsen, D.; Gribbestad, I.; Bathen, T. Principal component analysis for the comparison of metabolic profiles from human rectal cancer biopsies and colorectal xenografts using high-resolution magic angle spinning 1H magnetic resonance spectroscopy. Mol. Cancer 2008, 7 (1), 33. (34) Denkert, C.; Budczies, J.; Weichert, W.; Wohlgemuth, G.; Scholz, M.; Kind, T.; Niesporek, S.; Noske, A.; Buckendahl, A.; Dietel, M.; Fiehn, O. Metabolite profiling of human colon carcinoma deregulation of TCA cycle and amino acid turnover. Mol. Cancer 2008, 7 (1), 72. (35) Hakomori, S.-i. Tumor malignancy defined by aberrant glycosylation and sphingo(glyco)lipid metabolism. Cancer Res. 1996, 56 (23), 5309−5318. (36) Hirayama, A.; Kami, K.; Sugimoto, M.; Sugawara, M.; Toki, N.; Onozuka, H.; Kinoshita, T.; Saito, N.; Ochiai, A.; Tomita, M.; Esumi, H.; Soga, T. Quantitative metabolome profiling of colon and stomach cancer microenvironment by capillary electrophoresis time-of-flight mass spectrometry. Cancer Res. 2009, 69, 4918−4925.

(9) Cairns, R. A.; Harris, I. S.; Mak, T. W. Regulation of cancer cell metabolism. Nat. Rev. Cancer 2011, 11, 85−95. (10) Griffin, J. L.; Kauppinen, R. A. Tumor metabolomics in animal models of human cancer. J. Proteome Res. 2007, 6, 498−505. (11) Voskoglou-Nomikos, T.; Pater, J. L.; Seymour, L. Clinical predictive value of the in vitro cell line, human xenograft, and mouse allograft preclinical cancer models. Clin. Cancer Res. 2003, 9 (11), 4227− 4239. (12) Rocha, C. M.; Carrola, J.; Barros, A. S.; Gil, A. M.; Goodfellow, B. J.; Carreira, I. M.; Bernardo, J.; Gomes, A.; Sousa, V.; Carvalho, L.; Duarte, I. F. Metabolic signatures of lung cancer in biofluids: NMRbased metabonomics of blood plasma. J. Proteome Res. 2011, 10 (9), 4314−4324. (13) Dettmer, K.; Aronov, P. A.; Hammock, B. D. Mass spectrometrybased metabolomics. Mass Spectrom. Rev. 2007, 26, 51−78. (14) Smith, C. A.; O’Maille, G.; Want, E. J.; Qin, C.; Trauger, S. A.; Brandon, T. R.; Custodio, D. E.; Abagyan, R.; Siuzdak, G. METLIN: A metabolite mass spectral database. Ther. Drug Monit. 2005, 27, 747− 751. (15) Horai, H.; Arita, M.; Kanaya, S.; Nihei, Y.; Ikeda, T.; Suwa, K.; Ojima, Y.; Tanaka, K.; Tanaka, S.; Aoshima, K.; Oda, Y.; Kakazu, Y.; Kusano, M.; Tohge, T.; Matsuda, F.; Sawada, Y.; Hirai, M. Y.; Nakanishi, H.; Ikeda, K.; Akimoto, N.; Maoka, T.; Takahashi, H.; Ara, T.; Sakurai, N.; Suzuki, H.; Shibata, D.; Neumann, S.; Iida, T.; Tanaka, K.; Funatsu, K.; Matsuura, F.; Soga, T.; Taguchi, R.; Saiot, K.; Nishioka, T. MassBank: a public repository for sharing mass spectral data for life sciences. J. Mass Spectrom. 2010, 45 (7), 703−714. (16) Wishart, D. S.; Knox, C.; Guo, A. C.; Eisner, R.; Young, N.; Gautam, B.; Hau, D. D.; Psychogios, N.; Dong, E.; Bouatra, S.; Mandal, R.; Sinelnikov, I.; Xia, J.; Jia, L.; Cruz, J. A.; Lim, E.; Sobsey, C. A.; Shrivastava, S.; Huang, P.; Liu, P.; Fang, L.; Peng, J.; Fradette, R.; Cheng, D.; Tzur, D.; Clements, M.; Lewis, A.; De Souza, A.; Zuniga, A.; Dawe, M.; Xiong, Y.; Clive, D.; Greiner, R.; Nazyrova, A.; Shaykhutdinov, R.; Li, L.; Vogel, H. J.; Forsythe, I. HMDB: a knowledgebase for the human metabolome. Nucleic Acids Res. 2009, 37, D603−610. (17) Wang, Y.; Xiao, J.; Suzek, T. O.; Zhang, J.; Wang, J.; Bryant, S. H. PubChem: a public information system for analyzing bioactivities of small molecules. Nucleic Acids Res. 2009, 37 (suppl 2), W623−W633. (18) Knox, C.; Shrivastava, S.; Stothard, P.; Eisner, R.; Wishart, D. S. BioSpider: a web server for automating metabolome annotations. Pac. Symp. Biocomput. 2007, 145−156. (19) Schmelzer, K.; Fahy, E.; Subramaniam, S.; Dennis, E. A. The lipid maps initiative in lipidomics. Methods Enzymol. 2007, 432, 171−183. (20) Kanehisa, M.; Araki, M.; Goto, S.; Hattori, M.; Hirakawa, M.; Itoh, M.; Katayama, T.; Kawashima, S.; Okuda, S.; Tokimatsu, T.; Yamanishi, Y. KEGG for linking genomes to life and the environment. Nucleic Acids Res. 2008, 36 Database, D480−D484. (21) Lean, C. L.; Mackinnon, W. B.; Delikatny, E. J.; Whitehead, R. H.; Mountford, C. E. Cell-surface fucosylation and magnetic resonance spectroscopy characterization of human malignant colorectal cells. Biochemistry 1992, 31 (45), 11095−11105. (22) Wilkinson, R. W.; Odedra, R.; Heaton, S. P.; Wedge, S. R.; Keen, N. J.; Crafter, C.; Foster, J. R.; Brady, M. C.; Bigley, A.; Brown, E.; Byth, K. F.; Barrass, N. C.; Mundt, K. E.; Foote, K. M.; Heron, N. M.; Jung, F. H.; Mortlock, A. A.; Boyle, F. T.; Green, S. AZD1152, a selective inhibitor of aurora B kinase, inhibits human tumor xenograft growth by inducing apoptosis. Clin. Cancer Res. 2007, 13 (12), 3682−3688. (23) Rantalainen, M.; Cloarec, O.; Beckonert, O.; Wilson, I. D.; Jackson, D.; Tonge, R.; Rowlinson, R.; Rayner, S.; Nickson, J.; Wilkinson, R. W.; Mills, J. D.; Trygg, J.; Nicholson, J. K.; Holmes, E. Statistically integrated metabonomic−proteomic studies on a human prostate cancer xenograft model in mice. J. Proteome Res. 2006, 5 (10), 2642−2655. (24) Sangster, T.; Major, H.; Plumb, R.; Wilson, A. J.; Wilson, I. D. A pragmatic and readily implemented quality control strategy for HPLC− MS and GC−MS-based metabonomic analysis. Analyst 2006, 131 (10), 1075−1078. (25) Gika, H. G.; Theodoridis, G. A.; Wingate, J. E.; Wilson, I. D. Within-day reproducibility of an HPLC−MS-based method for G

dx.doi.org/10.1021/pr400260h | J. Proteome Res. XXXX, XXX, XXX−XXX