Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Development and Quantitative Evaluation of a High-Resolution Metabolomics Technology

Development and Quantitative Evaluation of a High-Resolution Metabolomics Technology Article pubs.acs.org/ac Development and Quantitative Evaluation of a High-Resolution Metabolomics Technology Xiaojing Liu, Zheng Ser, and Jason W Locasale* Division of Nutritional Sciences, Cornell University, Ithaca, New York 14853, United States * Supporting Information ABSTRACT: Recent advances in mass spectrometry have allowed for unprecedented characterization of human metabolism and its contribution to disease. Despite these advances, limitations in metabolomics technology remain. Here, we describe a metab- olomics strategy that consolidates several recent improvements in mass spectrometry technology. The platform involves a high- resolution Orbitrap mass spectrometer coupled to faster scanning speeds, allowing for polarity switching and improved ion optics resulting in enhanced sensitivity. When coupled to HILIC chromatography, we are able to quantify over 339 metabolites from an extract of HCT8 cells with a linear range of over 4 orders of magnitude in a single chromatographic run. These metabolites include diverse chemical classes ranging from amino acids to polar lipids. In addition, we also detect over 3000 additional potential metabolites present in mammalian cells. We applied this platform to characterize the metabolome of eight colorectal cancer cell lines and observed both commonalities and heterogeneities across their metabolic profiles when cells are grown in identical conditions. Together these results demonstrate that simultaneous profiling and quantitation of the human metabolome is feasible. dvances in mass spectrometry have allowed for the In light of these advances, the extent of capability that this A simultaneous measurement and quantitation of many current metabolomics technology could allow remains poorly 1−4 metabolites in defined biological conditions. These advances characterized. We developed a HRMS-based metabolomics in metabolomics have led to newfound insights into the role of platform using HPLC coupled to a heated ESI source (HESI), a metabolism in health and disease. For example, tumor cells are quadrupole mass filter, a curved ion trap (C-trap), and Fourier known to have dramatic alterations in the ability to uptake and transform-based OrbitrapTM mass analyzer. This instrument, metabolize nutrients, resulting in gross rewiring of the termed the Q-Exactive MS (QE-MS), has demonstrated many 5−10 metabolic network. Mass spectrometry has played an superior capabilities for quantitative and qualitative proteomics 21−24 instrumental role in defining these differences that are now applications, but its general utility for metabolomics being investigated for cancer treatment and prevention. applications has, to our knowledge, yet to be explored. We These metabolomic technologies have involved high- next considered an extensive assessment of its performance in performance liquid chromatography (HPLC) coupled to an both targeted and nontargeted applications by evaluating its electrospray ion (ESI) source and mass analyzer. Typically, the ability to detect and quantify metabolomics across a set of platforms have used a triple quadrupole mass analyzer and colorectal cancer cell lines. involve targeting a series of metabolites by monitoring the transitions from the selected precursor ion to a specific EXPERIMENTAL SECTION fragmentation ion of the precursor ion (multiple reaction 11,12 Materials. All cell lines were provided as a generous gift monitoring, MRM). Alternatively, instruments utilizing from Dr. Lewis Cantley’s laboratory. RPMI 1640 medium was high-resolution mass spectrometry (HRMS) tend to have purchased from Cellgro. Fetal Bovine Serum (FBS), penicillin, higher duty cycle times, leading to difficulties in quantita- 13−15 and streptomycin were purchased from Hyclone Laboratories. tion. An instrument that consolidates these capabilities Dialyzed FBS was obtained from Life Technologies. Optima- could allow for untargeted metabolite profiling with sufficient grade ammonium acetate, ammonium hydroxide, acetonitrile, scan speeds for quantitative, targeted analysis. Such an advance methanol, and water were purchased from Fisher Scientific. might overcome many of the limitations in both approaches. Scan speeds have also improved such that polarity switching is obtainable on these instruments, allowing for approximately a Received: November 26, 2013 2-fold expansion of the number of metabolites that can be Accepted: January 10, 2014 16−20 detected during single chromatographic runs. Published: January 10, 2014 © 2014 American Chemical Society 2175 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 1. Overview of polar metabolite analysis platform. (A) The platform for polar metabolomics using LC-QE-MS. In positive mode, positively charged ions (red dots) are sent to S-lens (ion focusing), a quadrupole (low-resolution mass filter), C-trap (ions accumulate here until the targeted number of ions is reached), and finally Orbitrap high-resolution (HR) mass analyzer, where mass to charge ratio (m/z) of each ion and corresponding retention time (R.T.) are recorded. Once positive ions are sent to the Orbitrap from C-trap, the electronic field polarity is reversed, and only negatively charged ions (blue dots) are delivered from the HESI probe. (B) The duty cycle time when instrument is operated in pos/neg switch full scan mode with resolution of 70000. The typical duty cycle is between 512 and 912 ms, depending on the C-trap injection time (IT). Cell Culture. All cell lines were first cultured in 10 cm and mobile phase B is acetonitrile. The linear gradient used is as follows: 0 min, 85% B; 1.5 min, 85% B, 5.5 min, 35% B; 10 min, dishes with full growth medium, which contains RPMI 1640, 10% FBS, 100 units/mL penicillin and 100 μg/mL 35% B, 10.5 min, 35% B, 14.5 min, 35% B, 15 min, 85% B, and streptomycin. Cells were grown in a 37 °C incubator with 20 min, 85% B. The flow rate was 0.15 mL/min from 0 to 10 5% CO . min and 15 to 20 min and 0.3 mL/min from 10.5 to 14.5 min. Mass Spectrometry. The QE-MS is equipped with a HESI Sample Preparation for Dynamic Range Studies. HCT probe, and the relevant parameters are as listed: heater 8 cells were grown in three 10 cm dishes with full growth temperature, 120 °C; sheath gas, 30; auxiliary gas, 10; sweep medium. When the cells reach 80% confluence, the media were gas, 3; spray voltage, 3.6 kV for the positive mode and 2.5 kV quickly removed, and the dish was placed on top of dry ice. for the negative mode. Capillary temperature was set at 320 °C, Three milliliters of extraction solvent was immediately added and S-lens was 55. A full scan range from 60 to 900 (m/z) was (80% methanol/water), and the dishes were then transferred to used. The resolution was set at 70000. The maximum injection the −80 °C freezer. The dishes were left for 15 min, and then time (max IT) was 200 ms with typical injection times around cells were scraped into extraction solvent on dry ice. The 50 ms. These settings resulted in a duty cycle of around 550 ms entirety of the solution was transferred to two 1.7 mL eppendorf tubes and centrifuged with the speed of 20000g to carry out scans in both the positive and negative modes. Automated gain control (AGC) was targeted at 3 × 10 ions. for 10 min at 4 °C. Here, cell metabolite extracts were prepared For MS/MS, the isolation width of the precursor was set at 2.5, from three separate dishes to make three biological replicates. HCD collision energy was 35%, and max IT is 100 ms. The The supernatant was then transferred to new eppendorf tubes and dried in a SpeedVac. The samples can also be dried under resolution and AGC were 35000 and 200000, respectively. Full scan with resolution at 35000 and IT of 100 ms) was run nitrogen gas. After drying, one tube of each sample was stored together with MS/MS. Customized mass calibration was in the −80 °C freezer as a backup, while the other one was performed before any sample analysis. reconstituted into 20 μL of water (LC−MS grade, Fisher High-Performance Liquid Chromatography. The Scientific). A serial dilution of triplicate samples from 10 cm HPLC (Ultimate 3000 UHPLC) is coupled to QE-MS Petri dish was done 5 times with a dilution factor of 6, ending (Thermo Scientific) for metabolite separation and detection. up with 6 different concentrations of samples. These samples An Xbridge amide column (100 × 2.1 mm i.d., 3.5 μm; Waters) represent the amount of metabolites extracted from 10 , 1.67 × 6 5 4 3 3 is employed for compound separation at room temperature. 10 , 2.78 × 10 , 4.63 × 10 , 7.72 × 10 , and 1.29 × 10 of cells, The mobile phase A is 20 mM ammonium acetate and 15 mM respectively. Since each concentration of sample was prepared ammonium hydroxide in water with 3% acetonitrile, pH 9.0, in triplicate, a total of 18 samples are analyzed in LC-QE-MS. 2176 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 2. LC-QE-MS data analysis workflow. (A) Workflow for quantitative targeted and untargeted metabolomics study. (B) Workflow for unknown polar metabolites identification and scoring. Abbreviation: Stds = Standards. Metabolite Extraction from Colorectal Cancer Cell r distribution was represented as a histogram using GraphPad Lines. Eight colorectal cancer cell lines were seeded in 6-well 6.0. 5 5 Quantile normalization, unsupervised hierarchical clustering plates at the density of 2 × 10 to 5 × 10 per well for 24 h. (Pearson, Spearman linkages), and heat map generation were Metabolites were extracted as described above, except that 1 carried out with the software Gene-e (Broad institute, http:// mL of extraction solvent was used, instead of 3 mL. Each www.broadinstitute.org/cancer/software/GENE-E/index. sample was dissolved into 20 μL of water, and 5 μL was html). The maximum fold change (Maxchange) calculation was injected to LC-QE-MS. The sequence of sample injections was carried out in the software package R. randomized so that the fluctuation in LC-QE-MS performance was evenly distributed across each sample. Peak Extraction. Raw data collected from the LC-QE-MS RESULTS were processed on Thermo Scientific, Sieve 2.0. Peak alignment Overview. We first developed a strategy that focuses on and detection were performed according to manufacturer measuring polar metabolites (Figure 1, panels A and B). A cold protocols. For a targeted metabolomics analysis, a frameseed methanol extraction method was used to minimize the 25,26 including 194 metabolites was used for targeted metabolites perturbation of metabolism in cultured cells. LC-HRMS analysis with data collected in positive mode, while a frame seed with positive and negative mode switching was employed to of 262 metabolites was used for negative mode, where m/z expand on the number of metabolites that can be accessed. To width is set at 10 ppm. For an untargeted metabolomics achieve high throughput, we considered a chromatography run analysis, the following parameter values were used to extract of 20 min. Both untargeted and targeted metabolomics studies untargeted components (pairs of m/z and R.T.): background were carried out with the data obtained from the workflow in signal-to-noise ratio, 3; minimum ion count, 1 × 10 ; minimum Figure 1A. Figure 2A describes LC−MS data processing scans across the peak, 5; m/z step, 10 ppm. procedures. For the untargeted analysis, neither pre-existing Statistical Analysis. To assess the linear range, targeted knowledge of metabolites to be measured nor heavy isotope metabolite data was filtered as follows: for each metabolite, if labeled standards (Stds) are required. After a component the lowest signal in all of the samples is less than 10 and extraction, the data are further filtered by using multiple criteria, meanwhile the highest signal is less than 10 , then this including the coefficient of variation (CV) within replicate metabolite is considered as below the detection limit; if the samples and the total MS intensity (integrated peak area). lowest signal is less than 10 but the highest signal is more than Finally, components of interest are selected for a database 4 3 10 then replace the low signal with 10 . Calculations were search based on the detected mass of the selected component performed in R computing language (www.r-project.org). The with a 10 ppm mass tolerance. For targeted metabolomics, the 2177 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 3. MS/MS of positive ions with m/z of 428.04. (A) The extracted ion chromatogram (EIC) of m/z of 428.03669 (in positive mode) with a mass certainty within 10 ppm. (B) The full MS/MS chromatography of ions with m/z of 428.04 ± 1.25. (C) The MS/MS spectrum. The exact mass of fragment ion is shown below the corresponding fragment ion. Figure 4. Dynamic range of QE-MS. (A) The total ion chromatogram (TIC) for positive mode for increasing numbers of cells used. (B) TIC for negative mode for increasing numbers of cells used. (C) The log -transformed intensity distribution of targeted metabolites in 3 × 10 of HCT8 cells. An average of n = 3 biological replicates are considered. (D) The relationship between coefficient of variation (CV) of triplicate samples and MS intensity. The box plot shows the 75th/25th percentile, and the bar represents the median. (E) Linear regression analysis of each metabolite. The number of metabolites with a given r value is shown. corresponding mass to charge ratio (m/z) and retention time used, (3) there exists a unique single peak in the EIC channel (R.T.) are used for peak extraction. and this peak does not contain any known isomers, or if there A comprehensive list of metabolites with theoretical m/z are known isomers, there are characteristic MS/MS fragments (both in positive and negative mode) was generated based on a to distinguish the isomers, and (4) authentic standards are recent study. This list was used to generate extracted ion injected to confirm the assignment. On the basis of these chromatography (EIC) from full scan data. A scoring system criteria, we generated a list of 262 metabolites in negative mode was established to evaluate confidence in the metabolite and 194 in positive mode, and the following targeted assignments (Figure 2B). A metabolite peak will gain a positive metabolomics data processing was based on this list. score under any of the following situations: (1) Ions are MS/MS Identification of Isomers. An example of an MS/ detected in more than one concentration of sample, (2) a MS-based resolution of isomers is shown in Figure 3. corresponding C peak is detected when a labeled extract is Adenosine diphosphate (ADP) and deoxyguanosine diphos- 2178 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 5. MS intensity distribution and clustering in eight cell lines. (A) MS intensity distributions of cell extracts of colorectal cancer cell lines. Box plots represent the 75th/25th percentile, and the bars represent the median MS intensity. MS intensity is log transformed. (B) MS intensity distribution as in (A) but with quantile normalization. (C) Heat map of Pearson clustering of MS intensity in eight cell lines. (D) Heat map as in (C) but with quantile normalization. (E) Heat map of Spearman ranking clustering of MS intensity in eight cell line. (F) Heat map as in (E) but with quantile normalization. The color code bar is applicable to each of (C−F). phate (dGDP) are not distinguishable in full scan mode (Figure extracted from 10 cells and first diluted 6-fold and then 3A), since the two molecules have exactly the same elemental followed by serial dilution resulting in extracts of differing composition and, as a result, the same m/z.MS/MS concentrations. The total ion chromatography (TIC) from fragmentation by HCD was done at a resolution of 35000. these 6 concentrations are shown in Figure 4 (panels A and B) MS/MS peak (Figure 3B) and EIC from full scan (Figure 3A) (here, the Y axis is normalized by the highest intensity in the have the same retention time. At this resolution, MS/MS sample). Figure 4C demonstrates the MS intensity range across spectrum has decent intensity and, meanwhile, a very small targeted metabolites. Figure 4D demonstrates a strong mass error (1.2 ppm for fragment with m/z = 136.06161), as correlation between CV within triplicate samples and the shown in Figure 3 (panels B and C). In the MS/MS spectrum corresponding MS intensity. As expected, the higher the MS intensity, the lower the measured CV, since a lower signal tends (Figure 3C), the fragment of m/z = 348.06980 is generated from the cleavage of a phosphate group, which is not to have more interference from ions with very close m/z values. characteristic, while the fragment of m/z = 136.06161 is For metabolites with MS intensities higher than 1 × 10 , the corresponding to adenine, which can only be generated from CV is within 7.8% (at 75th percentile), while for MS intensities ADP by cleavage of the ribose group. There is no m/z = less than 1 × 10 , CV varies to a larger extent (132.8% at the 152.05669 (guanine from dGDP) detected, so the peak at 8.03 75th percentile). Therefore, we defined an MS intensity of 1 × min is assigned as ADP. We further confirmed this assignment 10 as the noise level, and in Figure 4E, data are processed by comparison of ADP and dGDP in a QT of MS/MS further by imputing intensities lower than 1 × 10 with a value 28 3 spectrum from the Massbank database. of 1 × 10 , as described in the methods section. The linear Dynamic Range of Metabolite Quantitation. Having regression of MS intensity (the integrated peak area within the developed a combined metabolomics technology, we next defined retention time window of every m/z) and concen- sought to evaluate its quantitative abilities. Metabolites were trations is shown in 4E. The TIC increases as the concentration 2179 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 6. Targeted metabolomic profiling in eight cell lines. (A) CV distribution of metabolites measured in eight cell lines. (B) CV distribution as in (A) except that quantile normalized MS intensity values were used. (C) Maxchange (log transformed) distribution of targeted metabolites. (D) Maxchange distribution as in (C) but with quantile normalized MS intensity values. Abbreviation: Maxchange, the ratio of maximum and minimum MS intensity for every component across cell lines. of injected metabolites increases, and meanwhile a linear HCT116, NCI-H508, and SW48 are clustered using raw regression analysis of 5 concentrations (excluding the highest intensity values, while SW620 is separated from other cell lines saturated concentration) shows that more than 86% of when raw intensity is quantile normalized. However, when metabolites detected have r values larger than 0.85, implying Spearman correlations are used for the linkages, quantile that over 4 orders of magnitude, the relative mass intensity can normalization makes little difference, as expected (Figure 5, accurately reflect metabolite relative levels. The low r in the panels E and F). remaining 14% of metabolites was either because they were not To compare the metabolite profiling variations in different detected at low sample concentration or because they had a cell lines, the mean values of every three replicates are used to poor linear MS response. At a number of 10 cells, signals calculate CV and Maxchange (N = 8). Here Maxchange is tended to decrease due to strong ion suppression from the defined as the ratio of highest to lowest mean intensity biological matrix effect, and also the retention times shift due to observed across the cell line panel. As demonstrated in Figure 6 overloading of the analyte on the LC column. (panels A and B), 270 out of 375 metabolites have CVs less Metabolic Profiling of Colorectal Cancer Cell Lines. than 40%, and this number increases to 290 if the quantile The method described and discussed in Figures 1 and 2 was normalized values are used. There are 27 out of 375 metabolites then applied to study the metabolite profiles in eight colorectal with CVs larger than 100% if working with raw values, while cancer cell lines: SW620, SW480, HCT8, HT29, HCT116, this number decreases to 24 if data is quantile normalized. NCI-H508, SW48, and SW948. A list of 375 measured targeted When a Maxchange value is calculated (Figure 6, panels C and ions is included in Table 1 of the Supporting Information. Each D), there are 190 metabolites with Maxchange ≤ 2. After cell line was cultured in the same medium to avoid confounding quantile normalization, this number increases to 222. For effects on metabolism due to differences in nutrient availability. metabolites with Maxchange ≥ 32, the number slightly Each cell line was observed to have a different, albeit small, increases from 20 to 22 with quantile normalization. intensity range (Figure 5A), which can be removed with CV (within triplicates) and MS intensity distributions of quantile normalization (Figure 5B). An inspection of untargeted components extracted from the cell line SW620 data metabolite intensities was carried out using different clustering based on the parameters listed in the method section are plotted in Figure 7 (panels A and B). The number of algorithms (Figure 5, panels C, D, E, and F). For each representation, the columns represent different metabolites, components (with MS intensity higher than 10 ) extracted at while the rows represent eight cell lines in triplicate. The effect different cutoff values is plotted in Figure 7C. Here, the 10 MS of quantile normalization becomes apparent when clustering is intensity cutoff value is used to avoid working with massive carried out using linkages corresponding to Pearson correla- untargeted amounts of components data and to improve data tions. As shown in Figure 5 (panels C and D), SW620, HT29, quality. On the basis of Figure 7 (panels B and C), MS intensity 2180 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 7. Untargeted component extraction. (A) CV (within biology triplicates) distribution of untargeted components. (B) MS intensity distribution of untargeted components in cell line SW620. (C) The relationship between the number of extracted components and CV cutoff values. (D) MS intensity distributions of cell extract of colorectal cancer cell lines. Box plots represent the 75th/25th percentile, and the bar represents the median. In (A and B), there are no filters applied to the extracted components, while in (D), filters of MS intensity higher than 10 and CV (within triplicate) less than 20% were applied. higher than 104 and CV less than 20% are used to filter the raw derivatives and folates, which are not detected in our method. intensity data, and the filtered intensity range of each cell line is It is either due to their low abundance in the cells lines we used 29−31 plotted in Figure 7D. The untargeted components data show or their instability. Therefore, for these metabolites, similar trends as the targeted metabolite data (Figure 8 (panels additional optimization of the extraction procedure will be A and B). Quantile normalization increases the number of required. metabolites (Maxchange ≤ 2) from 1940 to 2099 and For a long time, MS was not considered as a quantitative meanwhile decreases the number of metabolites (Maxchange analytical technology, because for metabolites with different ≥ 32) from 34 to 20. When untargeted components intensities chemical structures, they tend to have different ionization are used, as shown in Figure 8 (panels C−F), neither Spearman efficiency, and even for the same metabolite, if it is measured at rank-based clustering nor Pearson correlation-based clustering different times or spiked into different biological samples, the is affected by quantile normalization of MS intensity. However, MS response tends to fluctuate, which is due to the matrix 32,33 compared to the clustering pattern based on the Spearman effect. Therefore, stable heavy isotope-labeled standards ranking of 375 targeted metabolites, clustering based on 2931 (stds) are commonly spiked into unknown samples to correct untargeted components is different for the cell line SW620. The the error introduced by sample preparations and MS response pool of metabolites tends to affect the clustering pattern for fluctuations. In our LC−MS setup, within a wide range, the both Pearson correlations and Spearman ranking based- MS intensity increases in a linear pattern when the clustering. However, there are few conserved subclusters, corresponding samples are prepared from increasing cell such as cell lines HT29 and HCT116 and cell lines HCT8 numbers, which gives us high confidence of label-free and SW948, which are always clustered together regardless of differential quantitative analysis based on our current workflow. the clustering method used. This linearity is observed even when samples of interest are randomly dispersed across large sample runs. Moreover, the DISCUSSION CVs within biological triplicates at sufficient peak intensity levels are very small, implying that our current workflow is very Our chromatography method involving HPLC, employed a reproducible, and subtle biological variations of metabolites in high pH mobile phase and amide column, coupled with different samples can be measured. However, too much positive/negative switching HRMS enables us to analyze both material results in severe ion suppression and also induces acidic and basic polar metabolites in a single experiment. Even overload in the LC, so overall, a smaller number of cells (3 × though this method is not optimized for recovery of any 5 6 10 to 2 × 10 of cells) results in a larger number of metabolites specific metabolite, it nevertheless enables us to cover a large number of polar metabolites and lipids. Moreover, there are capable of being detected. To overcome the day-to-day some important polar metabolites, such as coenzyme A variation (i.e., a batch effect), advanced statistic analysis 2181 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 8. Metabolomic profiling in eight cell lines. (A) Maxchange distribution of cell extract. (B) Maxchange distribution as in (A) except that quantile normalized MS intensity values were used. (C) Heat map of Pearson clustering of MS intensity in eight cell line. (D) Heat map as in (C) but with quantile normalization. (E) Heat map of Spearman ranking clustering of MS intensity in eight cell lines. (F) Heat map as in (E) but with quantile normalization. The color code bar is applicable to (C−F). Abbreviation: Maxchange, the ratio of maximum and minimum MS intensity for every component across cell lines. 34,35 might be helpful. For absolute quantitation, internal stds or to 2931 untargeted components, as shown in Figure 5 (panels external calibration curves are still required, although it is C−F) and Figure 8 (C−F). Therefore, in our study, a targeted conceivable that a regression model could circumvent the need approach is done first to make metabolic profiling, and then for calibration curves in some instances. very specific filters are applied to narrow down untargeted In order to compare the metabolomics profiling differences components, followed by database searching to make sure no in different cell types, quantile normalization was applied to interesting metabolite is missed. rescale the metabolite intensities. However, based on our study, MS/MS data can further increase confidence of unknown quantile normalization results in only modest effects on the CV metabolite identification, especially for metabolites with or Maxchange calculations. Its effect on clustering patterns isomers and poor separation on LC (Figure 3). However, the across the whole data set is however readily apparent. MS/MS database is far from complete, and also MS/MS Since HRMS records almost every ion falling into the scan spectra in the database were generated from different types of range and above the limit of detection, little effort is required to mass spectrometry with different fragmentation methods. It has build a detection method for each metabolite, but an efficient been shown that the MS/MS pattern is dependent on how approach to deal with massive data is critical. Untargeted collision energy is applied and also the elemental composition component-based approaches cover almost every ion recorded of the collision gas. Moreover, to obtain useful MS/MS in the spectrum if filtering parameters are set at very low values, spectra, a precursor ion of sufficient intensity is required. Due but this would be inefficient in its computational cost. to these limitations, efforts are needed to further develop a high Practically, a targeted approach is more efficient, even though throughput method for MS/MS data processing and metabolite metabolites outside of the list will be missed. In practice, in identification. order to compare metabolic profiling in different samples, our Eight colorectal cancer cell lines show three distinct current targeted list gave a similar cluster pattern as compared metabolic patterns which gives us a hint that the metabolic 2182 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article (9) Locasale, J. W.; Grassian, A. R.; Melman, T.; Lyssiotis, C. A.; enzymes are either differentially expressed or with variant Mattaini, K. R.; Bass, A. J.; Heffron, G.; Metallo, C. M.; Muranen, T.; activities across these cell lines. This potentially suggests Sharfi, H.; Sasaki, A. T.; Anastasiou, D.; Mullarky, E.; Vokes, N. I.; opportunities for biomarker analysis in metabolomics applica- Sasaki, M.; Beroukhim, R.; Stephanopoulos, G.; Ligon, A. H.; tions. Meyerson, M.; Richardson, A. L.; Chin, L.; Wagner, G.; Asara, J. M.; Brugge, J. S.; Cantley, L. C.; Vander Heiden, M. G. Nat. Genet. 2011, CONCLUSION 43, 869−874. The platform demonstrated here is applicable for targeted and (10) Locasale, J. W.; Melman, T.; Song, S.; Yang, X.; Swanson, K. D.; untargeted label-free polar metabolites quantitative analysis. Cantley, L. C.; Wong, E. T.; Asara, J. M. Mol. Cell. Proteomics 2012, 11, M111 014688. Besides cell culture work, this method is being applied to (11) Scherb, J.; Kreissl, J.; Haupt, S.; Schieberle, P. J. Agric. Food biomarker studies using tissues, serum, and other human fluids, Chem. 2009, 57, 9091−9096. and provides a resource to the metabolomics field. With such a (12) Ciccimaro, E.; Blair, I. A. Bioanalysis 2010, 2, 311−341. technology, further investigation that connects metabolite (13) Southam, A. D.; Payne, T. G.; Cooper, H. J.; Arvanitis, T. N.; profile to biological phenotype is possible. Viant, M. R. Anal. Chem. 2007, 79, 4595−4602. (14) Dettmer, K.; Aronov, P. A.; Hammock, B. D. Mass Spectrom. ASSOCIATED CONTENT ■ Rev. 2007, 26,51−78. * S Supporting Information (15) Lu, W.; Clasquin, M. F.; Melamud, E.; Amador-Noguez, D.; Caudy, A. A.; Rabinowitz, J. D. Anal. Chem. 2010, 82, 3212−3221. The authors declare no conflicts of interest. This material is (16) Michalski, A.; Damoc, E.; Hauschild, J. P.; Lange, O.; Wieghaus, available free of charge via the Internet at http://pubs.acs.org. A.; Makarov, A.; Nagaraj, N.; Cox, J.; Mann, M.; Horning, S. Mol. Cell. Proteomics 2011, 10, M111 011015. AUTHOR INFORMATION (17) Fedorova, G.; Randak, T.; Lindberg, R. H.; Grabic, R. Rapid Corresponding Author Commun. Mass Spectrom. 2013, 27, 1751−1762. *E-mail: [email protected]. (18) Blasco, H.; Corcia, P.; Pradat, P. F.; Bocca, C.; Gordon, P. H.; Veyrat-Durebex, C.; Mavel, S.; Nadal-Desbarats, L.; Moreau, C.; Notes Devos, D.; Andres, C. R.; Emond, P. J. Proteome Res. 2013, 12, 3746− The authors declare no competing financial interest. (19) Li, T. M.; Chen, J.; Li, X.; Ding, X. J.; Wu, Y.; Zhao, L. F.; Chen, ACKNOWLEDGMENTS S.; Lei, X.; Dong, M. Q. Anal. Chem. 2013, 85, 9281−9287. The authors would like to acknowledge Detlef Schumann and (20) Concheiro, M.; Lee, D.; Lendoiro, E.; Huestis, M. A. J. Chromatogr., A 2013, 1297, 123−130. Jennifer Sutton (Thermo Scientific) for valuable discussions on (21) Nagaraj, N.; Kulak, N. A.; Cox, J.; Neuhauser, N.; Mayr, K.; data processing. The authors would also like to thank Brandon Hoerning, O.; Vorm, O.; Mann, M. Mol. Cell. Proteomics 2012, 11, Barker, Tamar Melman. and Mahya Mehrmohamadi (Cornell M111 013722. University) for help with data analysis. Research reported in (22) Kelstrup, C. D.; Young, C.; Lavallee, R.; Nielsen, M. L.; Olsen, J. this publication was supported by the National Cancer Institute V. J. Proteome Res. 2012, 11, 3487−3497. of the National Institutes of Health under Award Number (23) Gallien, S.; Duriez, E.; Crone, C.; Kellmann, M.; Moehring, T.; R00CA168997. The content is solely the responsibility of the Domon, B. Mol. Cell. Proteomics 2012, 11, 1709−1723. authors and does not necessarily represent the official views of (24) Poulsen, J. W.; Madsen, C. T.; Young, C.; Kelstrup, C. D.; Grell, the National Institutes of Health. H. C.; Henriksen, P.; Juhl-Jensen, L.; Nielsen, M. L. J. Proteomics 2012, 75, 3886−3897. REFERENCES (25) Yuan, M.; Breitkopf, S. B.; Yang, X.; Asara, J. M. Nat. Protoc. 2012, 7, 872−881. (1) Denkert, C.; Budczies, J.; Kind, T.; Weichert, W.; Tablack, P.; (26) Shyh-Chang, N.; Locasale, J. W.; Lyssiotis, C. A.; Zheng, Y.; Sehouli, J.; Niesporek, S.; Konsgen, D.; Dietel, M.; Fiehn, O. Cancer Teo, R. Y.; Ratanasirintrawoot, S.; Zhang, J.; Onder, T.; Unternaehrer, Res. 2006, 66, 10795−10804. J. J.; Zhu, H.; Asara, J. M.; Daley, G. Q.; Cantley, L. C. Science 2013, (2) Smith, C. A.; Want, E. J.; O’Maille, G.; Abagyan, R.; Siuzdak, G. 339, 222−226. Anal. Chem. 2006, 78, 779−787. (27) Thiele, I.; Swainston, N.; Fleming, R. M.; Hoppe, A.; Sahoo, S.; (3) Cornett, D. S.; Reyzer, M. L.; Chaurand, P.; Caprioli, R. M. Nat. Aurich, M. K.; Haraldsdottir, H.; Mo, M. L.; Rolfsson, O.; Stobbe, M. Methods 2007, 4, 828−833. D.; Thorleifsson, S. G.; Agren, R.; Bolling, C.; Bordel, S.; Chavali, A. (4) Sreekumar, A.; Poisson, L. M.; Rajendiran, T. M.; Khan, A. P.; K.; Dobson, P.; Dunn, W. B.; Endler, L.; Hala, D.; Hucka, M.; Hull, D.; Cao, Q.; Yu, J.; Laxman, B.; Mehra, R.; Lonigro, R. J.; Li, Y.; Nyati, M. Jameson, D.; Jamshidi, N.; Jonsson, J. J.; Juty, N.; Keating, S.; K.; Ahsan, A.; Kalyana-Sundaram, S.; Han, B.; Cao, X.; Byun, J.; Nookaew, I.; Le Novere, N.; Malys, N.; Mazein, A.; Papin, J. A.; Price, Omenn, G. S.; Ghosh, D.; Pennathur, S.; Alexander, D. C.; Berger, A.; N. D.; Selkov, E., Sr.; Sigurdsson, M. I.; Simeonidis, E.; Sonnenschein, Shuster, J. R.; Wei, J. T.; Varambally, S.; Beecher, C.; Chinnaiyan, A. N.; Smallbone, K.; Sorokin, A.; van Beek, J. H.; Weichart, D.; M. Nature 2009, 457, 910−914. Goryanin, I.; Nielsen, J.; Westerhoff, H. V.; Kell, D. B.; Mendes, P.; (5) Griffin, J. L.; Shockcor, J. P. Nat. Rev. Cancer 2004, 4, 551−561. Palsson, B. O. Nat. Biotechnol. 2013, 31, 419−425. (6) Christofk, H. R.; Vander Heiden, M. G.; Harris, M. H.; (28) Horai, H.; Arita, M.; Kanaya, S.; Nihei, Y.; Ikeda, T.; Suwa, K.; Ramanathan, A.; Gerszten, R. E.; Wei, R.; Fleming, M. D.; Schreiber, S. Ojima, Y.; Tanaka, K.; Tanaka, S.; Aoshima, K.; Oda, Y.; Kakazu, Y.; L.; Cantley, L. C. Nature 2008, 452, 230−233. Kusano, M.; Tohge, T.; Matsuda, F.; Sawada, Y.; Hirai, M. Y.; (7) Dang, L.; White, D. W.; Gross, S.; Bennett, B. D.; Bittinger, M. Nakanishi, H.; Ikeda, K.; Akimoto, N.; Maoka, T.; Takahashi, H.; Ara, A.; Driggers, E. M.; Fantin, V. R.; Jang, H. G.; Jin, S.; Keenan, M. C.; T.; Sakurai, N.; Suzuki, H.; Shibata, D.; Neumann, S.; Iida, T.; Tanaka, Marks, K. M.; Prins, R. M.; Ward, P. S.; Yen, K. E.; Liau, L. M.; K.; Funatsu, K.; Matsuura, F.; Soga, T.; Taguchi, R.; Saito, K.; Rabinowitz, J. D.; Cantley, L. C.; Thompson, C. B.; Vander Heiden, M. G.; Su, S. M. Nature 2009, 462, 739−744. Nishioka, T. J. Mass Spectrom. 2010, 45, 703−714. (8) Vander Heiden, M. G.; Locasale, J. W.; Swanson, K. D.; Sharfi, (29) Ishikawa,M.; Fujino,T.; Sakashita, H.;Morikawa, K.; H.; Heffron, G. J.; Amador-Noguez, D.; Christofk, H. R.; Wagner, G.; Yamamoto, T. Tohoku J. Exp. Med. 1995, 175,55−67. Rabinowitz, J. D.; Asara, J. M.; Cantley, L. C. Science 2010, 329, 1492− (30) Garbis, S. D.; Melse-Boonstra, A.; West, C. E.; van Breemen, R. 1499. B. Anal. Chem. 2001, 73, 5358−5364. 2183 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article (31) Pfeiffer, C. M.; Fazili, Z.; McCoy, L.; Zhang, M.; Gunter, E. W. Clin. Chem. 2004, 50, 423−432. (32) Annesley, T. M. Clin. Chem. 2003, 49, 1041−1044. (33) Taylor, P. J. Clin. Biochem. 2005, 38, 328−334. (34) De Livera, A. M.; Dias, D. A.; De Souza, D.; Rupasinghe, T.; Pyke, J.; Tull, D.; Roessner, U.; McConville, M.; Speed, T. P. Anal. Chem. 2012, 84, 10768−10776. (35) De Livera, A. M.; Olshansky, M.; Speed, T. P. Methods Mol. Biol. 2013, 1055, 291−307. 2184 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Analytical Chemistry Pubmed Central

Development and Quantitative Evaluation of a High-Resolution Metabolomics Technology

Liu, Xiaojing; Ser, Zheng; Locasale, Jason W
Analytical Chemistry , Volume 86 (4) – Jan 10, 2014
Download PDF
10 pages

Loading next page...
 
/lp/pubmed-central/development-and-quantitative-evaluation-of-a-high-resolution-AgZxZEugZR

References (85)

Publisher
Pubmed Central
Copyright
Copyright © 2014 American Chemical Society
ISSN
0003-2700
eISSN
1520-6882
DOI
10.1021/ac403845u
Publisher site
See Article on Publisher Site

Abstract

Article pubs.acs.org/ac Development and Quantitative Evaluation of a High-Resolution Metabolomics Technology Xiaojing Liu, Zheng Ser, and Jason W Locasale* Division of Nutritional Sciences, Cornell University, Ithaca, New York 14853, United States * Supporting Information ABSTRACT: Recent advances in mass spectrometry have allowed for unprecedented characterization of human metabolism and its contribution to disease. Despite these advances, limitations in metabolomics technology remain. Here, we describe a metab- olomics strategy that consolidates several recent improvements in mass spectrometry technology. The platform involves a high- resolution Orbitrap mass spectrometer coupled to faster scanning speeds, allowing for polarity switching and improved ion optics resulting in enhanced sensitivity. When coupled to HILIC chromatography, we are able to quantify over 339 metabolites from an extract of HCT8 cells with a linear range of over 4 orders of magnitude in a single chromatographic run. These metabolites include diverse chemical classes ranging from amino acids to polar lipids. In addition, we also detect over 3000 additional potential metabolites present in mammalian cells. We applied this platform to characterize the metabolome of eight colorectal cancer cell lines and observed both commonalities and heterogeneities across their metabolic profiles when cells are grown in identical conditions. Together these results demonstrate that simultaneous profiling and quantitation of the human metabolome is feasible. dvances in mass spectrometry have allowed for the In light of these advances, the extent of capability that this A simultaneous measurement and quantitation of many current metabolomics technology could allow remains poorly 1−4 metabolites in defined biological conditions. These advances characterized. We developed a HRMS-based metabolomics in metabolomics have led to newfound insights into the role of platform using HPLC coupled to a heated ESI source (HESI), a metabolism in health and disease. For example, tumor cells are quadrupole mass filter, a curved ion trap (C-trap), and Fourier known to have dramatic alterations in the ability to uptake and transform-based OrbitrapTM mass analyzer. This instrument, metabolize nutrients, resulting in gross rewiring of the termed the Q-Exactive MS (QE-MS), has demonstrated many 5−10 metabolic network. Mass spectrometry has played an superior capabilities for quantitative and qualitative proteomics 21−24 instrumental role in defining these differences that are now applications, but its general utility for metabolomics being investigated for cancer treatment and prevention. applications has, to our knowledge, yet to be explored. We These metabolomic technologies have involved high- next considered an extensive assessment of its performance in performance liquid chromatography (HPLC) coupled to an both targeted and nontargeted applications by evaluating its electrospray ion (ESI) source and mass analyzer. Typically, the ability to detect and quantify metabolomics across a set of platforms have used a triple quadrupole mass analyzer and colorectal cancer cell lines. involve targeting a series of metabolites by monitoring the transitions from the selected precursor ion to a specific EXPERIMENTAL SECTION fragmentation ion of the precursor ion (multiple reaction 11,12 Materials. All cell lines were provided as a generous gift monitoring, MRM). Alternatively, instruments utilizing from Dr. Lewis Cantley’s laboratory. RPMI 1640 medium was high-resolution mass spectrometry (HRMS) tend to have purchased from Cellgro. Fetal Bovine Serum (FBS), penicillin, higher duty cycle times, leading to difficulties in quantita- 13−15 and streptomycin were purchased from Hyclone Laboratories. tion. An instrument that consolidates these capabilities Dialyzed FBS was obtained from Life Technologies. Optima- could allow for untargeted metabolite profiling with sufficient grade ammonium acetate, ammonium hydroxide, acetonitrile, scan speeds for quantitative, targeted analysis. Such an advance methanol, and water were purchased from Fisher Scientific. might overcome many of the limitations in both approaches. Scan speeds have also improved such that polarity switching is obtainable on these instruments, allowing for approximately a Received: November 26, 2013 2-fold expansion of the number of metabolites that can be Accepted: January 10, 2014 16−20 detected during single chromatographic runs. Published: January 10, 2014 © 2014 American Chemical Society 2175 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 1. Overview of polar metabolite analysis platform. (A) The platform for polar metabolomics using LC-QE-MS. In positive mode, positively charged ions (red dots) are sent to S-lens (ion focusing), a quadrupole (low-resolution mass filter), C-trap (ions accumulate here until the targeted number of ions is reached), and finally Orbitrap high-resolution (HR) mass analyzer, where mass to charge ratio (m/z) of each ion and corresponding retention time (R.T.) are recorded. Once positive ions are sent to the Orbitrap from C-trap, the electronic field polarity is reversed, and only negatively charged ions (blue dots) are delivered from the HESI probe. (B) The duty cycle time when instrument is operated in pos/neg switch full scan mode with resolution of 70000. The typical duty cycle is between 512 and 912 ms, depending on the C-trap injection time (IT). Cell Culture. All cell lines were first cultured in 10 cm and mobile phase B is acetonitrile. The linear gradient used is as follows: 0 min, 85% B; 1.5 min, 85% B, 5.5 min, 35% B; 10 min, dishes with full growth medium, which contains RPMI 1640, 10% FBS, 100 units/mL penicillin and 100 μg/mL 35% B, 10.5 min, 35% B, 14.5 min, 35% B, 15 min, 85% B, and streptomycin. Cells were grown in a 37 °C incubator with 20 min, 85% B. The flow rate was 0.15 mL/min from 0 to 10 5% CO . min and 15 to 20 min and 0.3 mL/min from 10.5 to 14.5 min. Mass Spectrometry. The QE-MS is equipped with a HESI Sample Preparation for Dynamic Range Studies. HCT probe, and the relevant parameters are as listed: heater 8 cells were grown in three 10 cm dishes with full growth temperature, 120 °C; sheath gas, 30; auxiliary gas, 10; sweep medium. When the cells reach 80% confluence, the media were gas, 3; spray voltage, 3.6 kV for the positive mode and 2.5 kV quickly removed, and the dish was placed on top of dry ice. for the negative mode. Capillary temperature was set at 320 °C, Three milliliters of extraction solvent was immediately added and S-lens was 55. A full scan range from 60 to 900 (m/z) was (80% methanol/water), and the dishes were then transferred to used. The resolution was set at 70000. The maximum injection the −80 °C freezer. The dishes were left for 15 min, and then time (max IT) was 200 ms with typical injection times around cells were scraped into extraction solvent on dry ice. The 50 ms. These settings resulted in a duty cycle of around 550 ms entirety of the solution was transferred to two 1.7 mL eppendorf tubes and centrifuged with the speed of 20000g to carry out scans in both the positive and negative modes. Automated gain control (AGC) was targeted at 3 × 10 ions. for 10 min at 4 °C. Here, cell metabolite extracts were prepared For MS/MS, the isolation width of the precursor was set at 2.5, from three separate dishes to make three biological replicates. HCD collision energy was 35%, and max IT is 100 ms. The The supernatant was then transferred to new eppendorf tubes and dried in a SpeedVac. The samples can also be dried under resolution and AGC were 35000 and 200000, respectively. Full scan with resolution at 35000 and IT of 100 ms) was run nitrogen gas. After drying, one tube of each sample was stored together with MS/MS. Customized mass calibration was in the −80 °C freezer as a backup, while the other one was performed before any sample analysis. reconstituted into 20 μL of water (LC−MS grade, Fisher High-Performance Liquid Chromatography. The Scientific). A serial dilution of triplicate samples from 10 cm HPLC (Ultimate 3000 UHPLC) is coupled to QE-MS Petri dish was done 5 times with a dilution factor of 6, ending (Thermo Scientific) for metabolite separation and detection. up with 6 different concentrations of samples. These samples An Xbridge amide column (100 × 2.1 mm i.d., 3.5 μm; Waters) represent the amount of metabolites extracted from 10 , 1.67 × 6 5 4 3 3 is employed for compound separation at room temperature. 10 , 2.78 × 10 , 4.63 × 10 , 7.72 × 10 , and 1.29 × 10 of cells, The mobile phase A is 20 mM ammonium acetate and 15 mM respectively. Since each concentration of sample was prepared ammonium hydroxide in water with 3% acetonitrile, pH 9.0, in triplicate, a total of 18 samples are analyzed in LC-QE-MS. 2176 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 2. LC-QE-MS data analysis workflow. (A) Workflow for quantitative targeted and untargeted metabolomics study. (B) Workflow for unknown polar metabolites identification and scoring. Abbreviation: Stds = Standards. Metabolite Extraction from Colorectal Cancer Cell r distribution was represented as a histogram using GraphPad Lines. Eight colorectal cancer cell lines were seeded in 6-well 6.0. 5 5 Quantile normalization, unsupervised hierarchical clustering plates at the density of 2 × 10 to 5 × 10 per well for 24 h. (Pearson, Spearman linkages), and heat map generation were Metabolites were extracted as described above, except that 1 carried out with the software Gene-e (Broad institute, http:// mL of extraction solvent was used, instead of 3 mL. Each www.broadinstitute.org/cancer/software/GENE-E/index. sample was dissolved into 20 μL of water, and 5 μL was html). The maximum fold change (Maxchange) calculation was injected to LC-QE-MS. The sequence of sample injections was carried out in the software package R. randomized so that the fluctuation in LC-QE-MS performance was evenly distributed across each sample. Peak Extraction. Raw data collected from the LC-QE-MS RESULTS were processed on Thermo Scientific, Sieve 2.0. Peak alignment Overview. We first developed a strategy that focuses on and detection were performed according to manufacturer measuring polar metabolites (Figure 1, panels A and B). A cold protocols. For a targeted metabolomics analysis, a frameseed methanol extraction method was used to minimize the 25,26 including 194 metabolites was used for targeted metabolites perturbation of metabolism in cultured cells. LC-HRMS analysis with data collected in positive mode, while a frame seed with positive and negative mode switching was employed to of 262 metabolites was used for negative mode, where m/z expand on the number of metabolites that can be accessed. To width is set at 10 ppm. For an untargeted metabolomics achieve high throughput, we considered a chromatography run analysis, the following parameter values were used to extract of 20 min. Both untargeted and targeted metabolomics studies untargeted components (pairs of m/z and R.T.): background were carried out with the data obtained from the workflow in signal-to-noise ratio, 3; minimum ion count, 1 × 10 ; minimum Figure 1A. Figure 2A describes LC−MS data processing scans across the peak, 5; m/z step, 10 ppm. procedures. For the untargeted analysis, neither pre-existing Statistical Analysis. To assess the linear range, targeted knowledge of metabolites to be measured nor heavy isotope metabolite data was filtered as follows: for each metabolite, if labeled standards (Stds) are required. After a component the lowest signal in all of the samples is less than 10 and extraction, the data are further filtered by using multiple criteria, meanwhile the highest signal is less than 10 , then this including the coefficient of variation (CV) within replicate metabolite is considered as below the detection limit; if the samples and the total MS intensity (integrated peak area). lowest signal is less than 10 but the highest signal is more than Finally, components of interest are selected for a database 4 3 10 then replace the low signal with 10 . Calculations were search based on the detected mass of the selected component performed in R computing language (www.r-project.org). The with a 10 ppm mass tolerance. For targeted metabolomics, the 2177 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 3. MS/MS of positive ions with m/z of 428.04. (A) The extracted ion chromatogram (EIC) of m/z of 428.03669 (in positive mode) with a mass certainty within 10 ppm. (B) The full MS/MS chromatography of ions with m/z of 428.04 ± 1.25. (C) The MS/MS spectrum. The exact mass of fragment ion is shown below the corresponding fragment ion. Figure 4. Dynamic range of QE-MS. (A) The total ion chromatogram (TIC) for positive mode for increasing numbers of cells used. (B) TIC for negative mode for increasing numbers of cells used. (C) The log -transformed intensity distribution of targeted metabolites in 3 × 10 of HCT8 cells. An average of n = 3 biological replicates are considered. (D) The relationship between coefficient of variation (CV) of triplicate samples and MS intensity. The box plot shows the 75th/25th percentile, and the bar represents the median. (E) Linear regression analysis of each metabolite. The number of metabolites with a given r value is shown. corresponding mass to charge ratio (m/z) and retention time used, (3) there exists a unique single peak in the EIC channel (R.T.) are used for peak extraction. and this peak does not contain any known isomers, or if there A comprehensive list of metabolites with theoretical m/z are known isomers, there are characteristic MS/MS fragments (both in positive and negative mode) was generated based on a to distinguish the isomers, and (4) authentic standards are recent study. This list was used to generate extracted ion injected to confirm the assignment. On the basis of these chromatography (EIC) from full scan data. A scoring system criteria, we generated a list of 262 metabolites in negative mode was established to evaluate confidence in the metabolite and 194 in positive mode, and the following targeted assignments (Figure 2B). A metabolite peak will gain a positive metabolomics data processing was based on this list. score under any of the following situations: (1) Ions are MS/MS Identification of Isomers. An example of an MS/ detected in more than one concentration of sample, (2) a MS-based resolution of isomers is shown in Figure 3. corresponding C peak is detected when a labeled extract is Adenosine diphosphate (ADP) and deoxyguanosine diphos- 2178 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 5. MS intensity distribution and clustering in eight cell lines. (A) MS intensity distributions of cell extracts of colorectal cancer cell lines. Box plots represent the 75th/25th percentile, and the bars represent the median MS intensity. MS intensity is log transformed. (B) MS intensity distribution as in (A) but with quantile normalization. (C) Heat map of Pearson clustering of MS intensity in eight cell lines. (D) Heat map as in (C) but with quantile normalization. (E) Heat map of Spearman ranking clustering of MS intensity in eight cell line. (F) Heat map as in (E) but with quantile normalization. The color code bar is applicable to each of (C−F). phate (dGDP) are not distinguishable in full scan mode (Figure extracted from 10 cells and first diluted 6-fold and then 3A), since the two molecules have exactly the same elemental followed by serial dilution resulting in extracts of differing composition and, as a result, the same m/z.MS/MS concentrations. The total ion chromatography (TIC) from fragmentation by HCD was done at a resolution of 35000. these 6 concentrations are shown in Figure 4 (panels A and B) MS/MS peak (Figure 3B) and EIC from full scan (Figure 3A) (here, the Y axis is normalized by the highest intensity in the have the same retention time. At this resolution, MS/MS sample). Figure 4C demonstrates the MS intensity range across spectrum has decent intensity and, meanwhile, a very small targeted metabolites. Figure 4D demonstrates a strong mass error (1.2 ppm for fragment with m/z = 136.06161), as correlation between CV within triplicate samples and the shown in Figure 3 (panels B and C). In the MS/MS spectrum corresponding MS intensity. As expected, the higher the MS intensity, the lower the measured CV, since a lower signal tends (Figure 3C), the fragment of m/z = 348.06980 is generated from the cleavage of a phosphate group, which is not to have more interference from ions with very close m/z values. characteristic, while the fragment of m/z = 136.06161 is For metabolites with MS intensities higher than 1 × 10 , the corresponding to adenine, which can only be generated from CV is within 7.8% (at 75th percentile), while for MS intensities ADP by cleavage of the ribose group. There is no m/z = less than 1 × 10 , CV varies to a larger extent (132.8% at the 152.05669 (guanine from dGDP) detected, so the peak at 8.03 75th percentile). Therefore, we defined an MS intensity of 1 × min is assigned as ADP. We further confirmed this assignment 10 as the noise level, and in Figure 4E, data are processed by comparison of ADP and dGDP in a QT of MS/MS further by imputing intensities lower than 1 × 10 with a value 28 3 spectrum from the Massbank database. of 1 × 10 , as described in the methods section. The linear Dynamic Range of Metabolite Quantitation. Having regression of MS intensity (the integrated peak area within the developed a combined metabolomics technology, we next defined retention time window of every m/z) and concen- sought to evaluate its quantitative abilities. Metabolites were trations is shown in 4E. The TIC increases as the concentration 2179 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 6. Targeted metabolomic profiling in eight cell lines. (A) CV distribution of metabolites measured in eight cell lines. (B) CV distribution as in (A) except that quantile normalized MS intensity values were used. (C) Maxchange (log transformed) distribution of targeted metabolites. (D) Maxchange distribution as in (C) but with quantile normalized MS intensity values. Abbreviation: Maxchange, the ratio of maximum and minimum MS intensity for every component across cell lines. of injected metabolites increases, and meanwhile a linear HCT116, NCI-H508, and SW48 are clustered using raw regression analysis of 5 concentrations (excluding the highest intensity values, while SW620 is separated from other cell lines saturated concentration) shows that more than 86% of when raw intensity is quantile normalized. However, when metabolites detected have r values larger than 0.85, implying Spearman correlations are used for the linkages, quantile that over 4 orders of magnitude, the relative mass intensity can normalization makes little difference, as expected (Figure 5, accurately reflect metabolite relative levels. The low r in the panels E and F). remaining 14% of metabolites was either because they were not To compare the metabolite profiling variations in different detected at low sample concentration or because they had a cell lines, the mean values of every three replicates are used to poor linear MS response. At a number of 10 cells, signals calculate CV and Maxchange (N = 8). Here Maxchange is tended to decrease due to strong ion suppression from the defined as the ratio of highest to lowest mean intensity biological matrix effect, and also the retention times shift due to observed across the cell line panel. As demonstrated in Figure 6 overloading of the analyte on the LC column. (panels A and B), 270 out of 375 metabolites have CVs less Metabolic Profiling of Colorectal Cancer Cell Lines. than 40%, and this number increases to 290 if the quantile The method described and discussed in Figures 1 and 2 was normalized values are used. There are 27 out of 375 metabolites then applied to study the metabolite profiles in eight colorectal with CVs larger than 100% if working with raw values, while cancer cell lines: SW620, SW480, HCT8, HT29, HCT116, this number decreases to 24 if data is quantile normalized. NCI-H508, SW48, and SW948. A list of 375 measured targeted When a Maxchange value is calculated (Figure 6, panels C and ions is included in Table 1 of the Supporting Information. Each D), there are 190 metabolites with Maxchange ≤ 2. After cell line was cultured in the same medium to avoid confounding quantile normalization, this number increases to 222. For effects on metabolism due to differences in nutrient availability. metabolites with Maxchange ≥ 32, the number slightly Each cell line was observed to have a different, albeit small, increases from 20 to 22 with quantile normalization. intensity range (Figure 5A), which can be removed with CV (within triplicates) and MS intensity distributions of quantile normalization (Figure 5B). An inspection of untargeted components extracted from the cell line SW620 data metabolite intensities was carried out using different clustering based on the parameters listed in the method section are plotted in Figure 7 (panels A and B). The number of algorithms (Figure 5, panels C, D, E, and F). For each representation, the columns represent different metabolites, components (with MS intensity higher than 10 ) extracted at while the rows represent eight cell lines in triplicate. The effect different cutoff values is plotted in Figure 7C. Here, the 10 MS of quantile normalization becomes apparent when clustering is intensity cutoff value is used to avoid working with massive carried out using linkages corresponding to Pearson correla- untargeted amounts of components data and to improve data tions. As shown in Figure 5 (panels C and D), SW620, HT29, quality. On the basis of Figure 7 (panels B and C), MS intensity 2180 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 7. Untargeted component extraction. (A) CV (within biology triplicates) distribution of untargeted components. (B) MS intensity distribution of untargeted components in cell line SW620. (C) The relationship between the number of extracted components and CV cutoff values. (D) MS intensity distributions of cell extract of colorectal cancer cell lines. Box plots represent the 75th/25th percentile, and the bar represents the median. In (A and B), there are no filters applied to the extracted components, while in (D), filters of MS intensity higher than 10 and CV (within triplicate) less than 20% were applied. higher than 104 and CV less than 20% are used to filter the raw derivatives and folates, which are not detected in our method. intensity data, and the filtered intensity range of each cell line is It is either due to their low abundance in the cells lines we used 29−31 plotted in Figure 7D. The untargeted components data show or their instability. Therefore, for these metabolites, similar trends as the targeted metabolite data (Figure 8 (panels additional optimization of the extraction procedure will be A and B). Quantile normalization increases the number of required. metabolites (Maxchange ≤ 2) from 1940 to 2099 and For a long time, MS was not considered as a quantitative meanwhile decreases the number of metabolites (Maxchange analytical technology, because for metabolites with different ≥ 32) from 34 to 20. When untargeted components intensities chemical structures, they tend to have different ionization are used, as shown in Figure 8 (panels C−F), neither Spearman efficiency, and even for the same metabolite, if it is measured at rank-based clustering nor Pearson correlation-based clustering different times or spiked into different biological samples, the is affected by quantile normalization of MS intensity. However, MS response tends to fluctuate, which is due to the matrix 32,33 compared to the clustering pattern based on the Spearman effect. Therefore, stable heavy isotope-labeled standards ranking of 375 targeted metabolites, clustering based on 2931 (stds) are commonly spiked into unknown samples to correct untargeted components is different for the cell line SW620. The the error introduced by sample preparations and MS response pool of metabolites tends to affect the clustering pattern for fluctuations. In our LC−MS setup, within a wide range, the both Pearson correlations and Spearman ranking based- MS intensity increases in a linear pattern when the clustering. However, there are few conserved subclusters, corresponding samples are prepared from increasing cell such as cell lines HT29 and HCT116 and cell lines HCT8 numbers, which gives us high confidence of label-free and SW948, which are always clustered together regardless of differential quantitative analysis based on our current workflow. the clustering method used. This linearity is observed even when samples of interest are randomly dispersed across large sample runs. Moreover, the DISCUSSION CVs within biological triplicates at sufficient peak intensity levels are very small, implying that our current workflow is very Our chromatography method involving HPLC, employed a reproducible, and subtle biological variations of metabolites in high pH mobile phase and amide column, coupled with different samples can be measured. However, too much positive/negative switching HRMS enables us to analyze both material results in severe ion suppression and also induces acidic and basic polar metabolites in a single experiment. Even overload in the LC, so overall, a smaller number of cells (3 × though this method is not optimized for recovery of any 5 6 10 to 2 × 10 of cells) results in a larger number of metabolites specific metabolite, it nevertheless enables us to cover a large number of polar metabolites and lipids. Moreover, there are capable of being detected. To overcome the day-to-day some important polar metabolites, such as coenzyme A variation (i.e., a batch effect), advanced statistic analysis 2181 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article Figure 8. Metabolomic profiling in eight cell lines. (A) Maxchange distribution of cell extract. (B) Maxchange distribution as in (A) except that quantile normalized MS intensity values were used. (C) Heat map of Pearson clustering of MS intensity in eight cell line. (D) Heat map as in (C) but with quantile normalization. (E) Heat map of Spearman ranking clustering of MS intensity in eight cell lines. (F) Heat map as in (E) but with quantile normalization. The color code bar is applicable to (C−F). Abbreviation: Maxchange, the ratio of maximum and minimum MS intensity for every component across cell lines. 34,35 might be helpful. For absolute quantitation, internal stds or to 2931 untargeted components, as shown in Figure 5 (panels external calibration curves are still required, although it is C−F) and Figure 8 (C−F). Therefore, in our study, a targeted conceivable that a regression model could circumvent the need approach is done first to make metabolic profiling, and then for calibration curves in some instances. very specific filters are applied to narrow down untargeted In order to compare the metabolomics profiling differences components, followed by database searching to make sure no in different cell types, quantile normalization was applied to interesting metabolite is missed. rescale the metabolite intensities. However, based on our study, MS/MS data can further increase confidence of unknown quantile normalization results in only modest effects on the CV metabolite identification, especially for metabolites with or Maxchange calculations. Its effect on clustering patterns isomers and poor separation on LC (Figure 3). However, the across the whole data set is however readily apparent. MS/MS database is far from complete, and also MS/MS Since HRMS records almost every ion falling into the scan spectra in the database were generated from different types of range and above the limit of detection, little effort is required to mass spectrometry with different fragmentation methods. It has build a detection method for each metabolite, but an efficient been shown that the MS/MS pattern is dependent on how approach to deal with massive data is critical. Untargeted collision energy is applied and also the elemental composition component-based approaches cover almost every ion recorded of the collision gas. Moreover, to obtain useful MS/MS in the spectrum if filtering parameters are set at very low values, spectra, a precursor ion of sufficient intensity is required. Due but this would be inefficient in its computational cost. to these limitations, efforts are needed to further develop a high Practically, a targeted approach is more efficient, even though throughput method for MS/MS data processing and metabolite metabolites outside of the list will be missed. In practice, in identification. order to compare metabolic profiling in different samples, our Eight colorectal cancer cell lines show three distinct current targeted list gave a similar cluster pattern as compared metabolic patterns which gives us a hint that the metabolic 2182 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article (9) Locasale, J. W.; Grassian, A. R.; Melman, T.; Lyssiotis, C. A.; enzymes are either differentially expressed or with variant Mattaini, K. R.; Bass, A. J.; Heffron, G.; Metallo, C. M.; Muranen, T.; activities across these cell lines. This potentially suggests Sharfi, H.; Sasaki, A. T.; Anastasiou, D.; Mullarky, E.; Vokes, N. I.; opportunities for biomarker analysis in metabolomics applica- Sasaki, M.; Beroukhim, R.; Stephanopoulos, G.; Ligon, A. H.; tions. Meyerson, M.; Richardson, A. L.; Chin, L.; Wagner, G.; Asara, J. M.; Brugge, J. S.; Cantley, L. C.; Vander Heiden, M. G. Nat. Genet. 2011, CONCLUSION 43, 869−874. The platform demonstrated here is applicable for targeted and (10) Locasale, J. W.; Melman, T.; Song, S.; Yang, X.; Swanson, K. D.; untargeted label-free polar metabolites quantitative analysis. Cantley, L. C.; Wong, E. T.; Asara, J. M. Mol. Cell. Proteomics 2012, 11, M111 014688. Besides cell culture work, this method is being applied to (11) Scherb, J.; Kreissl, J.; Haupt, S.; Schieberle, P. J. Agric. Food biomarker studies using tissues, serum, and other human fluids, Chem. 2009, 57, 9091−9096. and provides a resource to the metabolomics field. With such a (12) Ciccimaro, E.; Blair, I. A. Bioanalysis 2010, 2, 311−341. technology, further investigation that connects metabolite (13) Southam, A. D.; Payne, T. G.; Cooper, H. J.; Arvanitis, T. N.; profile to biological phenotype is possible. Viant, M. R. Anal. Chem. 2007, 79, 4595−4602. (14) Dettmer, K.; Aronov, P. A.; Hammock, B. D. Mass Spectrom. ASSOCIATED CONTENT ■ Rev. 2007, 26,51−78. * S Supporting Information (15) Lu, W.; Clasquin, M. F.; Melamud, E.; Amador-Noguez, D.; Caudy, A. A.; Rabinowitz, J. D. Anal. Chem. 2010, 82, 3212−3221. The authors declare no conflicts of interest. This material is (16) Michalski, A.; Damoc, E.; Hauschild, J. P.; Lange, O.; Wieghaus, available free of charge via the Internet at http://pubs.acs.org. A.; Makarov, A.; Nagaraj, N.; Cox, J.; Mann, M.; Horning, S. Mol. Cell. Proteomics 2011, 10, M111 011015. AUTHOR INFORMATION (17) Fedorova, G.; Randak, T.; Lindberg, R. H.; Grabic, R. Rapid Corresponding Author Commun. Mass Spectrom. 2013, 27, 1751−1762. *E-mail: [email protected]. (18) Blasco, H.; Corcia, P.; Pradat, P. F.; Bocca, C.; Gordon, P. H.; Veyrat-Durebex, C.; Mavel, S.; Nadal-Desbarats, L.; Moreau, C.; Notes Devos, D.; Andres, C. R.; Emond, P. J. Proteome Res. 2013, 12, 3746− The authors declare no competing financial interest. (19) Li, T. M.; Chen, J.; Li, X.; Ding, X. J.; Wu, Y.; Zhao, L. F.; Chen, ACKNOWLEDGMENTS S.; Lei, X.; Dong, M. Q. Anal. Chem. 2013, 85, 9281−9287. The authors would like to acknowledge Detlef Schumann and (20) Concheiro, M.; Lee, D.; Lendoiro, E.; Huestis, M. A. J. Chromatogr., A 2013, 1297, 123−130. Jennifer Sutton (Thermo Scientific) for valuable discussions on (21) Nagaraj, N.; Kulak, N. A.; Cox, J.; Neuhauser, N.; Mayr, K.; data processing. The authors would also like to thank Brandon Hoerning, O.; Vorm, O.; Mann, M. Mol. Cell. Proteomics 2012, 11, Barker, Tamar Melman. and Mahya Mehrmohamadi (Cornell M111 013722. University) for help with data analysis. Research reported in (22) Kelstrup, C. D.; Young, C.; Lavallee, R.; Nielsen, M. L.; Olsen, J. this publication was supported by the National Cancer Institute V. J. Proteome Res. 2012, 11, 3487−3497. of the National Institutes of Health under Award Number (23) Gallien, S.; Duriez, E.; Crone, C.; Kellmann, M.; Moehring, T.; R00CA168997. The content is solely the responsibility of the Domon, B. Mol. Cell. Proteomics 2012, 11, 1709−1723. authors and does not necessarily represent the official views of (24) Poulsen, J. W.; Madsen, C. T.; Young, C.; Kelstrup, C. D.; Grell, the National Institutes of Health. H. C.; Henriksen, P.; Juhl-Jensen, L.; Nielsen, M. L. J. Proteomics 2012, 75, 3886−3897. REFERENCES (25) Yuan, M.; Breitkopf, S. B.; Yang, X.; Asara, J. M. Nat. Protoc. 2012, 7, 872−881. (1) Denkert, C.; Budczies, J.; Kind, T.; Weichert, W.; Tablack, P.; (26) Shyh-Chang, N.; Locasale, J. W.; Lyssiotis, C. A.; Zheng, Y.; Sehouli, J.; Niesporek, S.; Konsgen, D.; Dietel, M.; Fiehn, O. Cancer Teo, R. Y.; Ratanasirintrawoot, S.; Zhang, J.; Onder, T.; Unternaehrer, Res. 2006, 66, 10795−10804. J. J.; Zhu, H.; Asara, J. M.; Daley, G. Q.; Cantley, L. C. Science 2013, (2) Smith, C. A.; Want, E. J.; O’Maille, G.; Abagyan, R.; Siuzdak, G. 339, 222−226. Anal. Chem. 2006, 78, 779−787. (27) Thiele, I.; Swainston, N.; Fleming, R. M.; Hoppe, A.; Sahoo, S.; (3) Cornett, D. S.; Reyzer, M. L.; Chaurand, P.; Caprioli, R. M. Nat. Aurich, M. K.; Haraldsdottir, H.; Mo, M. L.; Rolfsson, O.; Stobbe, M. Methods 2007, 4, 828−833. D.; Thorleifsson, S. G.; Agren, R.; Bolling, C.; Bordel, S.; Chavali, A. (4) Sreekumar, A.; Poisson, L. M.; Rajendiran, T. M.; Khan, A. P.; K.; Dobson, P.; Dunn, W. B.; Endler, L.; Hala, D.; Hucka, M.; Hull, D.; Cao, Q.; Yu, J.; Laxman, B.; Mehra, R.; Lonigro, R. J.; Li, Y.; Nyati, M. Jameson, D.; Jamshidi, N.; Jonsson, J. J.; Juty, N.; Keating, S.; K.; Ahsan, A.; Kalyana-Sundaram, S.; Han, B.; Cao, X.; Byun, J.; Nookaew, I.; Le Novere, N.; Malys, N.; Mazein, A.; Papin, J. A.; Price, Omenn, G. S.; Ghosh, D.; Pennathur, S.; Alexander, D. C.; Berger, A.; N. D.; Selkov, E., Sr.; Sigurdsson, M. I.; Simeonidis, E.; Sonnenschein, Shuster, J. R.; Wei, J. T.; Varambally, S.; Beecher, C.; Chinnaiyan, A. N.; Smallbone, K.; Sorokin, A.; van Beek, J. H.; Weichart, D.; M. Nature 2009, 457, 910−914. Goryanin, I.; Nielsen, J.; Westerhoff, H. V.; Kell, D. B.; Mendes, P.; (5) Griffin, J. L.; Shockcor, J. P. Nat. Rev. Cancer 2004, 4, 551−561. Palsson, B. O. Nat. Biotechnol. 2013, 31, 419−425. (6) Christofk, H. R.; Vander Heiden, M. G.; Harris, M. H.; (28) Horai, H.; Arita, M.; Kanaya, S.; Nihei, Y.; Ikeda, T.; Suwa, K.; Ramanathan, A.; Gerszten, R. E.; Wei, R.; Fleming, M. D.; Schreiber, S. Ojima, Y.; Tanaka, K.; Tanaka, S.; Aoshima, K.; Oda, Y.; Kakazu, Y.; L.; Cantley, L. C. Nature 2008, 452, 230−233. Kusano, M.; Tohge, T.; Matsuda, F.; Sawada, Y.; Hirai, M. Y.; (7) Dang, L.; White, D. W.; Gross, S.; Bennett, B. D.; Bittinger, M. Nakanishi, H.; Ikeda, K.; Akimoto, N.; Maoka, T.; Takahashi, H.; Ara, A.; Driggers, E. M.; Fantin, V. R.; Jang, H. G.; Jin, S.; Keenan, M. C.; T.; Sakurai, N.; Suzuki, H.; Shibata, D.; Neumann, S.; Iida, T.; Tanaka, Marks, K. M.; Prins, R. M.; Ward, P. S.; Yen, K. E.; Liau, L. M.; K.; Funatsu, K.; Matsuura, F.; Soga, T.; Taguchi, R.; Saito, K.; Rabinowitz, J. D.; Cantley, L. C.; Thompson, C. B.; Vander Heiden, M. G.; Su, S. M. Nature 2009, 462, 739−744. Nishioka, T. J. Mass Spectrom. 2010, 45, 703−714. (8) Vander Heiden, M. G.; Locasale, J. W.; Swanson, K. D.; Sharfi, (29) Ishikawa,M.; Fujino,T.; Sakashita, H.;Morikawa, K.; H.; Heffron, G. J.; Amador-Noguez, D.; Christofk, H. R.; Wagner, G.; Yamamoto, T. Tohoku J. Exp. Med. 1995, 175,55−67. Rabinowitz, J. D.; Asara, J. M.; Cantley, L. C. Science 2010, 329, 1492− (30) Garbis, S. D.; Melse-Boonstra, A.; West, C. E.; van Breemen, R. 1499. B. Anal. Chem. 2001, 73, 5358−5364. 2183 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184 Analytical Chemistry Article (31) Pfeiffer, C. M.; Fazili, Z.; McCoy, L.; Zhang, M.; Gunter, E. W. Clin. Chem. 2004, 50, 423−432. (32) Annesley, T. M. Clin. Chem. 2003, 49, 1041−1044. (33) Taylor, P. J. Clin. Biochem. 2005, 38, 328−334. (34) De Livera, A. M.; Dias, D. A.; De Souza, D.; Rupasinghe, T.; Pyke, J.; Tull, D.; Roessner, U.; McConville, M.; Speed, T. P. Anal. Chem. 2012, 84, 10768−10776. (35) De Livera, A. M.; Olshansky, M.; Speed, T. P. Methods Mol. Biol. 2013, 1055, 291−307. 2184 dx.doi.org/10.1021/ac403845u | Anal. Chem. 2014, 86, 2175−2184

Journal

Analytical Chemistry – Pubmed Central

Published: Jan 10, 2014

There are no references for this article.