Metabolomics pp 141-155 | Cite as

Application of Liquid Chromatography-Mass Spectrometry Analysis in Metabolomics

Reversed-Phase Monolithic Capillary Chromatography and Hydrophilic Chromatography Coupled to Electrospray Ionization-Mass Spectrometry
  • Vladimir V. Tolstikov
  • Oliver Fiehn
  • Nobuo Tanaka
Part of the Methods in Molecular Biology™ book series (MIMB, volume 358)

Abstract

Analysis of the entire metabolome as the sum of all detectable components in the sample rather than analysis of each individual metabolite is performed by the metabolomics approaches. To monitor in parallel hundreds or even thousands of metabolites, high-throughput techniques are required that enable screening for relative changes rather than absolute concentrations of compounds. Most analytical techniques for profiling small molecules consist of gas chromatography (GC) or high-performance liquid chromatography (HPLC) coupled to mass spectrometry. HPLC separations are better suited for the analysis of labile and high molecular weight compounds, and for the analysis of nonvolatile polar compounds in their natural form. Although GC- and HPLC-based profiling techniques are not truly quantitative, the compounds detecting and employing the acceptable standards may compare their relative amounts. We have demonstrated that reversed-phase monolithic capillary chromatography and hydrophilic chromatography can be successfully applied for sufficient plant crude extracts separations and metabolomics studies.

1 Introduction

In accordance with the central doctrine of molecular biology, DNA is transcribed into RNA then translated to proteins, which then make small molecules. However, often feedback loops and signaling circuits are overlooked, which may force the viewing of small molecules as actors and proteins (and DNA) as responding to alterations in metabolic levels. After all, the vision of complex regulatory networks will be closer to reality than a simplistic hierarchical paradigm. Although there may be over tens of thousands of genes, several hundred thousand transcripts, and up to 1 million proteins, it is estimated that there may be as few as 2000–3000 small molecules in the metabolome of higher organisms. The analysis of the metobolome looks very attractive from this point of view with fewer numbers of analytes to be identified and quantified. High chemical complexity, analytical and biological variance, and large dynamic range are quite challenging, even for the latest analytical methods. In most cases, analytical methods are based on chromatographic separation techniques like GC and high-performance liquid chromatography (HPLC), and in many cases comprise Fourier transform infrared spectroscopy, electron impact ionization-mass spectrometry (EI-MS), electrospray ionization-mass spectrometry (ESI-MS), and nuclear magnetic resonance (NMR) spectroscopy. Mass spectrometers are generally more sensitive and more selective than any other type of detector. Prior to MS detection, the metabolites have to be separated, and separated compounds must be ionized. Ionization techniques may vary, especially for GC-MS and LC-MS couplings. The high-throughput screening with GC- and LC-MS techniques generates large volumes of analytical data that require advanced software for data mining. Metabolomics studies and analysis of the tissue crude extracts cannot be accomplished with the use of a single separation/detection method owing to the high chemical diversity of the analyzed mixture. Hydrophobic components are nicely separated with the use of reversed-phase (RP) chromatography (1, 2, 3), which is very popular and an appreciated method of separation. Hydrophilic and charged small molecules are well separated by the capillary electrophoresis. Hydrophilic and neutral compounds are best suited for hydrophilic (HILIC) separation (4, 5, 6). We introduced HILIC-ESI-MS analysis for plant-derived samples and demonstrated feasibility of this approach, especially for analysis of the samples taken from hydrophilic compartments like the plant transport system (7, 8, 9, 10, 11). Traditional particles packed columns as well as high-speed monolithic columns can be used for HPLC separations prior to MS detection. Because the metabolomic approach requires large batches of the samples to be analyzed in order to apply statistical methods of the data treatment, micro-HPLC and capillary columns should be mostly used in order to avoid a significant amount of organic solvent evaporated into the atmosphere during this process. Silica-based C18-modified capillary monolithic columns actually offer a new step in micro-HPLC RP chromatography providing up to hundreds of thousands of theoretical plates per column. Therefore, performance of these columns is expected to be superior to conventional ones at the same range of HPLC parameters applied. We introduced the utilization of these columns for metabolomics studies and demonstrated their suitability for very complex mixture separations.

2 Materials

2.1 Standards and Chemicals

  1. 1.

    Oligosaccharides kit, l-amino acids kit, and reserpine are purchased from SAF (Taufkirchen and Seelze, Germany).

     
  2. 2.

    Ammonium acetate and acetic acid: highest purity grade available from SAF.

     

2.2 Solvents and Standards (seeNotes 1and2)

  1. 1.

    LC-MS-grade solvents are purchased from SAF.

     
  2. 2.

    Reserpine stock solution (0.2 mg/mL methanol).

     

2.3 Instrumentation

The LC-MS system consists of a Finnigan LCQ DECA mass spectrometer (ThermoFinnigan, San Jose, CA), a Rheos 2000 pump (Flux Instruments AB, Karlskoga, Sweden), and an HTS PAL auto sampler (CTC Analytics, Zwingen, Switzerland). The system was operated under the Xcalibur software (v1.3, ThermoFinnigan). Helium collision gas incoming pressure is kept at 2.6 bars, and the ion gauge pressure at 0.89 × 10−5. Full-scan mass spectra are acquired from 150 to 2000 amu at unit mass resolution. For MSn experiments, data-dependent scans are chosen with the wideband activation turned off. The normalized collision energy is set to 35%, and the activation Q is set to 0.250 with the source fragmentation turned off. Metal needle tips or online pico tips are from New Objective Inc., Woburn, MA. The mass spectrometer is tuned on sucrose solution (0.1 mg/mL) mixed with the acetonitrile/ammonium acetate buffer, pH 5.5 (1:1 [v/v]) prior to measurements.

2.3.1 HPLC-MS Setup

2.3.1.1 Normal Flow Setup

Normal flow HPLC operations with the use of conventional HPLC columns, having 4.6-mm I.D., require pneumatically assisted electrospray, and a postcolumn splitter that diverts in the range of 50 to 150 µL/min into the standard ESI ion source. Narrow bore columns do not require splitter. Nitrogen sheath gas pressure is set to six bars at the flow rate of 0.8 L/min. Spray voltage is set to 5 kV. The temperature of the heated transfer capillary is maintained at 250°C.

2.3.1.2 Micro Flow Setup

Micro-HPLC utilizing capillary columns does not require sheath gas. Precolumn splitter (6) is used with the split ratio in the range of 1:10 to 1:1000. Flow does not exceed 15 µL/min. Modified Protana nanospray source is used for column/tip assembly. A column is connected to a tip via a PEEK T-connector, with an inserted platinum wire in contact with the liquid. Voltage is applied through this wire. Spray voltage is set to 3 kV for positive and 2 kV for negative ionization modes. The temperature of the heated transfer capillary is maintained at 180°C.

2.4 Monolithic Silica-Based Capillary Columns (seeNote 3)

RP C18 monolithic silica-based capillary columns are manufactured in the laboratory of Professor Nobuo Tanaka, Kyoto Institute of Technology, Kyoto, Japan. Columns with the dimensions of 0.2-mm I.D. and 600 mm in length are used for the separations (10,11).

3 Methods

3.1 Sample Preparation for LC-MS Metabolomics Analysis

Sample preparation for LC-MS analysis is a very important part of the whole process. Live tissue or organ must be frozen prior to sampling because response to the sampling procedure quickly alters metabolism. Afterwards, extraction should be almost complete because an absence or insufficient amount of the component in the sample gives a zero detector response. This may not correspond to real tissue contents. Proteins should be precipitated and removed unless they are of particular interest. Concentration should be high enough to allow sufficient column loading with a low injection volume.

3.1.1 Leaf Harvest

  1. 1.

    Place a metal ball in each of the 2-mL sample tubes prior to harvest.

     
  2. 2.

    Submerge tubes into liquid nitrogen (N2).

     
  3. 3.

    Harvest 150 ± 30 mg of fresh weight of the plant leaves (Arabidopsis thaliana, Columbia 24) in sample tubes. Keep it in there and proceed with extraction.

     
  4. 4.

    Alternatively, perform rapid lyophilization of your tissue to obtain approx 15 mg dry weight. Tissue stored at −80°C for longer than 4 wk should not be used, as it was observed that considerable metabolic changes occur after this time.

     

3.1.2 Leaf Tissue Extraction

  1. 1.

    Prechill Retsch ball-mill tube holders in liquid N2. Put four samples into each of the ball-mill tube holders and homogenize the tissue for 1 min at 60% speed. Immediately put the tube holders and the samples back into liquid N2.

     
  2. 2.

    Alternatively, grind the tissue in liquid N2 using a mortar and pestle, and so on. Take out the sample tubes one-by-one and immediately add 1 mL of methanol in order to stop enzymatic activity. Vortex thoroughly. Add 50 µL of a reserpine stock solution (0.2 mg/mL methanol) as internal reference.

     
  3. 3.

    Add 50 µL of water and vortex. Shake the resulting suspension for 15 min at ambient temperature.

     
  4. 4.

    Centrifuge at 14,000g for 5 min. Carefully transfer the green supernatant into a sample glass vial that is equipped with a screw cap with Teflonized inlay.

     
  5. 5.

    Alternatively, acetonitrile or a mixture of isopropanol and acetone (1:1 [v/v]) can be used because rapid chlorophyll degradation occurs in methanol solutions during the storage.

     

3.1.3 Pumpkin Phloem Excudates

Pumpkin (giant pumpkin, Cucurbitacea maxima) phloem sampling should be done according to Richardson (12). Phloem samples are acquired from the fully expanded, mature leaves that did not show any signs of senescence in 8-wk-old plants.
  1. 1.

    In order to preserve water-soluble components intact and simultaneously stop enzymatic activity, dilute 100 µL of the freshly collected phloem exudates in 300 µL of pure water, and add 300 µL of chloroform to precipitate proteins by vortexing.

     
  2. 2.

    Collect the water phase and do rapid lyophilization.

     
  3. 3.

    Redissolve the residue in 50 µL of water/acetonitrile (1:1 [v/v]) mixture.

     
  4. 4.

    Centrifuge at 14,000g for 5 min. Carefully transfer the clear supernatant into a sample glass vial that is equipped with a screw cap with Teflonized inlay.

     

3.2 HILIC ESI-LC-MS Analyses of Pumpkin Phloem Excudates

Analytical LC is performed using acetonitrile (A) and 6.5 mM ammonium acetate (pH 5.5, adjusted by acetic acid) (B) as the mobile phase at the flow rates of 0.2–0.1 mL/min at the ambient temperature. LC-MS analysis is performed on TSK Gel Amide 80 column, 250 × 2.0 mm, 5-µm particle size (TosoHaas, Montgomeryville, PA). After 5 min of isocratic run at 0% B, gradient to 15% B is concluded at 10 min, then gradient to 55% B is completed at 80 min.
  1. 1.

    Prepare standards mix using oligosaccharides and L-amino acid kits in a mixture of acetonitrile:water (1:1 [v/v]). All the standards presented in these kits can be used in a single mix. For a simple chromatogram one can use a small number of standards in a mix. Concentration should not exceed 1 mg/mL.

     
  2. 2.

    Equilibrate column with the starting buffer for at least 20 min. Inject 10 µL and acquire the data in the full-scan mode for the positive and negative ions in the range of 100 to 1500 amu.

     
  3. 3.

    After acquisition finishes, wash the column with buffer B for 5 min and equilibrate for 20 min with buffer A prior the next run.

     
  4. 4.

    α-Amino acids are mostly detected as [M+H]+ positive ions. Mono- and oligosaccharides are detected as ammonia adducts in the positive mode and as [M-H]− ions in the negative mode. Hydrophobic α-amino acids are eluted earlier than basic and acidic ones. Oligosaccharides are eluted in the order of increasing monomer units. Larger oligomers are eluted latest.

     
  5. 5.

    Use selected standards as internal or external ones for the instrument calibration by the serial dilutions. This procedure is essential for further semi-quantitative analysis.

     
  6. 6.

    Prepare the pumpkin phloem excudate sample in accordance with Subheading 3.1.3. Inject 10 µL and acquire the data in the full-scan mode for the positive and negative ions.

     
  7. 7.

    Introduce selected internal standards and repeat analysis.

     
  8. 8.

    Refer to refs.2, 3, 4 for peaks annotation.

     
  9. 9.

    Include MS/MS or MSn experiments for both positive and negative modes in the analytical run in order to annotate peaks through the MS/MS libraries search and/or collect fragmentation information for de novo identification.

     
A typical HILIC LC-ESI-MS chromatogram of the pumpkin phloem excudate (C. maxima) is shown in Fig. 1. Some structural annotations are illustrated. Structural elucidation for components shown in Fig. 1 was accomplished by the comparison of the MSn fragmentation patterns and spectral data for authentic compounds available commercially and/or received from research laboratories where these substances have been isolated, characterized, and this data published. Unknown compounds, including oligomers, have been successfully isolated by subsequent fractions collection. Off-line nano-ESI-MSn, FT-ICRMS exact mass measurements, and two-dimensional NMR techniques have been applied to assign their chemical structures (4,7,8).
Fig. 1.

Hydrophilic liquid chromatography-electrospray ionization-mass spectrometry chromatograms of the pumpkin pholoem excudate (Cucurbitacea maxima). Peak annotations are given by the chemical structures.

3.3 HILIC Capillary ESI-LC-MS Analyses of the Plant Leaf Extracts

Split analytical LC is performed with the same HPLC pump and injector as for convenient chromatography using acetonitrile (A) and 6.5 mM ammonium acetate (pH 5.5, adjusted by acetic acid) (B) as mobile phase at pump flow rates of 0.15–0.06 mL/min at ambient temperature. LC-MS analysis is performed on polyhydroxyethyl A column, 150 × 0.6 mm, 3-µm particle size (PolyLC, Inc., Columbia, MD). Split ratio is set to 1:10. After 5 min of isocratic run at 0% B, gradient to 8% B is concluded at 5 min, then gradient to 35% B is completed at 90 min.
  1. 1.

    Prepare the plant leaf extract sample in accordance with Subheading 3.1.2. Equilibrate column with the starting buffer for at least 20 min. Inject 3 µL and acquire the data in the full-scan mode for the positive and negative ions in the range of 100 to 1500 amu.

     
  2. 2.

    After acquisition finishes, wash the column with buffer B for 10 min and equilibrate for 20 min with buffer A prior to the next run.

     
  3. 3.

    Use selected standards as internal or external ones for the instrument calibration by serial dilutions. This procedure is essential for further semi-quantitative analysis.

     
  4. 4.

    Introduce selected internal standards and repeat analysis.

     
  5. 5.

    Refer to Note 4 and publications (5, 6, 7, 8, 9,12) for peaks annotation.

     
  6. 6.

    Include MS/MS or MSn experiments for both positive and negative modes in the analytical run in order to annotate peaks through the MS/MS libraries search and/or collect fragmentation information for de novo identification.

     
Typical HILIC LC-ESI-MS chromatograms of plant leaf extracts (A. thaliana, Columbia 24, and Oriza sativa) are shown in Fig. 2.
Fig. 2.

Hydrophilic liquid chromatography-electrospray ionization-mass spectrometry chromatograms of the plant leaf extracts. (A) Chromatogram of Arabidopsis thaliana (Columbia 24) and (B)Oriza sativa. Peak annotations are given by the chemical classes.

3.4 RP C18 Capillary Monolithic ESI-LC-MS Analyses of Plant Leaf Extracts

Split analytical LC is performed using acetonitrile (B) and 6.5 mM ammonium acetate (pH 5.5, adjusted by acetic acid) (A) as the mobile phase at the pump flow rates of 0.05–0.2 mL/min at ambient temperature. LC-MS analysis is performed on a C18 monolithic silica-based column, 600 × 0.2 mm. Split ratio is set to 1:100. After 2 min of isocratic run at 0% B, gradient to 8% B is concluded at 5 min, then gradient to 30% B is completed at 25 min. Correspondingly, gradient to 70% B is completed at 35 min, then gradient to 99% B is completed at 50 min, and 99% B is run isocratically up to 100 min (seeNote 5) (Table 1).
  1. 1.

    Prepare plant leaf extract sample in accordance with Subheading 3.1.2. Equilibrate column with the starting buffer for at least 15 min. Inject 3 µL and acquire data in the full-scan mode for the positive and negative ions in the range of 100 to 1500 amu.

     
  2. 2.

    After acquisition finishes, wash the column with buffer B for 10 min and equilibrate for 15 min with buffer A prior to the next run.

     
  3. 3.

    Use selected standards (i.e., reserpine) as internal or external ones for the instrument calibration by the serial dilutions. This procedure is essential for further semi-quantitative analysis.

     
  4. 4.

    Introduce selected internal standards and repeat analysis.

     
  5. 5.

    Refer to Note 4 and refs.5, 6, 7, 8, 9,12 for peaks annotation.

     
  6. 6.

    Include MS/MS or MSn experiments for both positive and negative modes in the analytical run in order to annotate peaks through the MS/MS libraries search and/or collect fragmentation information for de novo identification.

     
Table 1

Gradient Flow Table for HPLC

 

Time (min)

%B

%A

HPLC flow rate (mL/min)

Startup

0.0

0

100

0.05

 

5.0

0

100

0.05

Run

0.0

0

100

0.05

 

2.0

0

100

0.05

 

5.0

8

92

0.05

 

25.0

30

70

0.07

 

35.0

70

30

0.08

 

50.0

99

1

0.10

 

100.0

100

0

0.15

Shut-down

120.0

100

0

0.20

 

125.0

0

100

0.07

 

130.0

0

100

0.05

 

150.0

0

100

0.05

Typical RP C18 monolithic LC-ESI-MS chromatograms of plant leaf extracts (A. thaliana, Columbia 24, and O. sativa) are shown in Fig. 3.
Fig. 3.

Reversed-phase C18 monolithic liquid chromatography-electrospray ionization-mass spectrometry (MS) chromatograms of plant leaf extracts. (A)Arabidopsis thaliana leaf extract. (B)Oriza sativa leaf extract. Examples for specific extracted ion chromatograms are located in the upper left corners. Compound identifications by MS/MS spectra and corresponding structures are situated in the upper right corners.

4 Notes

  1. 1.

    Unless stated otherwise, water-based buffers for HPLC should be refreshed daily to avoid mold formation and possible contamination owing to bacteria growth.

     
  2. 2.

    Each lot of acetonitrile should be investigated on the presence of admixtures by infusion into the mass spectrometer. Manufacturers provide purity control only by GC-MS. Deionized water quality should be as high as possible.

     
  3. 3.

    Monolithic silica capillary columns commercially available from Merck KGaA (Darmstadt, Germany), are prepared from tetramethoxysilane and available with 0.1-mm I.D. and up to 150-mm length. Performance of monolithic silica capillary columns including the tetramethoxysilane type can be optimized by splitting injection (as well as on-column detection in the case of ultraviolet detection), as described in refs.13 and 14.

     
  4. 4.

    Peak finding and deconvolution, utilizing software designed to handle large files and large datasets, can unravel large numbers of components, illustrated in Fig. 4. The number of components actually depends on the extraction protocol, which was recently demonstrated with the GC-MS and LC-MS methods (2). Structure elucidation and assignment, with the use of an ion trap mass spectrometer providing the fragmentation pathway, gets uncomplicated when the source of substance is known and fragments generated are in agreement with the predicted ones (Mass Frontier, HighChem), as illustrated in Fig. 5. Compound-specific fragmentation generating unique fragments actually provides a signature for this particular compound. Unfortunately it is problematic with the interpretation and in many cases requires strong evidence, like accurate masses and two-dimensional NMR data.

     
  5. 5.

    In gradient HPLC, increasing the stronger eluent (mobile phase B) can optimize separation by controlling a gradient range [ϕf (B% final)-ϕ0 (B% initial)], gradient time (tG), and flow rate (F), as well as column length. Three parameters, N (a number of theoretical plates of a column), α (a separation factor between peaks), and k (a retention factor of a solute) should be considered as in isocratic HPLC, particularly the average retention factor k* (Eq. 1, the retention factor at the middle of a column) for gradient elution (13, 14, 15, 16, 17). The k* is a function of S (the slope of the plot of log k against the content of solvent B in the mobile phase, S = 4∼10 for small molecules less than 1000 molecular weight (MW), S = ca. 30 for molecules with 10,000 MW, and S = ca. 100 for macromolecules with 100,000 MW, Vm (mobile phase volume in a column is given in µL), Δϕ (change in B solvent content in mobile phase from ϕ0 to ϕf: Δϕ = 1.0 for gradient from 0 to 100% B), F (µL/min), and tG (min) (13, 14, 15, 16, 17). as illustrated in Fig. 6.

     
Fig. 4.

Reversed-phase C18 monolithic liquid chromatography-electrospray ionization-mass spectrometry (MS) chromatograms of Arabidopsis thaliana leaf extract. (A) Components finding with the assistance of AnalyzerPro (Spectral Works) software. (B) Deconvoluted chromatogram with the ACS MS Manager suite (Advanced Chemistry Development).

Fig. 5.

(A) Structural assignment of the monomer. (B) Fragmentation pathways generated with the use Mass Frontier and MS2 spectra of two substances having different retention times.

Fig. 6.

Dependence of S value on the molecular weight of a solute. S value is a slope of the plots of log k values (retention factors of solutes) against φ (volume fraction of solvent B in mobile phase in isocratic elution).

In the case of the linear gradient of acetonitrile from 0 to 100%, k* is estimated to be ca. 5 for monolithic silica capillary columns of 60-cm lengths, 200-µm I.D. with tG = 100 min, and F = 4 µL/min. With fixed tG and F, the shorter column provides the greater k*. In the case of gradient elution of a macromolecule, a large S value results in a small k* value. As a consequence, a short column becomes advantageous. Because a greater flow rate will reduce the number of theoretical plates, it generally leads to the decrease in resolution.

When one optimizes the separation conditions, (1) k* should be kept around 5, and (2) a gradient range, Δϕ, should be minimized judging from the elution range of solutes in a given chromatogram, followed by (3) tuning of tG, F, and column lengths, if possible. Longer tG and a longer column lead to better resolution, whereas smaller k* and shorter tG lead to less resolution and better detection sensitivity, and smaller F results in better detection sensitivity. A good compromise between an efficient separation and good detection sensitivity can be obtained at k* = 5∼10 (13, 14, 15, 16, 17).

A certain level of a flow rate is required in order to provide a smooth gradient of the mobile phase. However, the higher flow rate leads to the greater dilution of solute bands, which will result in the decrease in detection sensitivity. Although the use of a longer column (increase in Vm) keeping k* and F values constant may lead to the increase in separation time and peak widths, improvement of separation efficiencies owing to the increase in the number of theoretical plates, and decrease in ion suppression, are expected (1). Because solutes are eluted with similar peak widths in linear gradient elution (Δϕ/tG = constant), it will be easy to optimize peak width for MS analysis by changing tG and column lengths.

References

  1. 1.
    Tolstikov, V. V., Lommen, A., Nakanishi, K., Tanaka, N., and Fiehn O. (2003) Monolithic silica-based capillary reversed phase liquid chromatography/electrospray mass spectrometry for plant metabolomics. Anal. Chem. 75, 6737–6740.PubMedCrossRefGoogle Scholar
  2. 2.
    Tolstikov, V. V., Fiehn, O., and Shulaev, V. (2005) The influence of the extraction methodology on global LC-MS and GC-MS metabolite profiling of arabidopsis thaliana leaf tissues. 53rd ASMS Conference, San Antonio, TX, June 5–9, 2005.Google Scholar
  3. 3.
    Tolstikov, V. V., Lorence, A., Cortes, D. F., et al. (2005) Use of capillary monolithic columns for untargeted metabolite profiling of the mutant lines overexpressing Miox4, the gene involved in L-ascorbic acid biosynthesis in arabidopsis. 28th International Symposium On Capillary Chromatography and Electrophoresis. Las Vegas, NV, May 22–25, 2005.Google Scholar
  4. 4.
    Tolstikov, V. V. and Fiehn, O. (2002) Analysis of highly polar compounds of plant origin: combination of hydrophilic interaction chromatography and electrospray ion trap mass spectrometry. Anal. Biochem. 301, 298–307.PubMedCrossRefGoogle Scholar
  5. 5.
    Tolstikov, V. V., Tanaka, N., and Fiehn O. (2003) Comprehensive metabolome analysis of crude Arabidopsis thaliana leaf extracts by LC/ESI-MSn/UV coupling. 2nd Plant Metabolomics Conference, Potsdam, Germany, April 25–28, 2003.Google Scholar
  6. 6.
    Tolstikov, V. V., Tanaka, N., and Fiehn, O. (2003) Metabolomics: LC-MS analysis development. Joint BTS/Cereal Chemistry Symposium, Adelaide, Australia, September 8–10, 2003.Google Scholar
  7. 7.
    Tolstikov, V. V., Zhang, B., Weckwerth, W., and Fiehn O. (2001) Structural investigation of O-glycans derived from plant material by the use of the HILIC HPLC separation and ESI-mass spectrometry. 49th ASMS conference on Mass Spectrometry and Allied Topics, Chicago, IL, May 27–31, 2001.Google Scholar
  8. 8.
    Tolstikov, V. V., Costisella, B., Weckwerth W., Zhang B., and Fiehn O. (2002) Accurate QTOF and MSn Ion trap measurements require additional NMR data for plant metabolites de-novo identification. 50th ASMS Conference on Mass Spectrometry and Allied Topics, Orlando, FL, June 2–7, 2002.Google Scholar
  9. 9.
    Tolstikov, V. V., Tanaka, N., and Fiehn, O. (2003) LC/MS analysis and development for plant metabolomic studies. 2003 LCMS Montreux Symposium. Savannah, GA, October 15–17, 2003.Google Scholar
  10. 10.
    Tanaka, N., Tolstikov, V., Weckwerth, W., Fiehn, O., and Fukusaki, H., (2003) Micro HPLC for metabolomics. In: Frontier of Metabolomic Research, Springer-Verlag, Tokyo, Japan, pp. 85–100.Google Scholar
  11. 11.
    Ikegami, T., Kobayashi, H., Kimura, H., Tolstikov, V., Fiehn, O., and Tanaka, N. (2005) HPLC for metabolomics: high efficiency separations utilizing monolithic silica columns. In: Metabolomics. The Frontiers of Systems Biology, Springer Verlag, Tokyo, Japan, pp. 107–126.Google Scholar
  12. 12.
    Nikiforova, V., Kopka, J., Tolstikov, V., Fiehn, O., Hesse, H., and Hoefgen, R. (2005) Systems rebalancing of metabolism in response to sulfur deprivation, as revealed by metabolome analysis of arabidopsis plants. Plant Physiology 138, 304–318.PubMedCrossRefGoogle Scholar
  13. 13.
    Kobayashi, H., Kajiwara, W., Inui, Y., et al. (2004) Chromatographic properties of monolithic silica capillary columns for polar and nonpolar compounds in reversed-phase HPLC. Chromatographia 60, S19–S25.CrossRefGoogle Scholar
  14. 14.
    Ikegami, T., Dicks, E., Kobayashi, H., et al. (2004) How to utilize true performance of monolithic silica columns? J. Sep. Sci. 27, 1292–1302.PubMedCrossRefGoogle Scholar
  15. 15.
    Richardson, P. T. and Baker, D. A. (1982) The chemical composition of cucurbit vascular exudates. J. Exp. Bot. 33, 1239–1247.CrossRefGoogle Scholar
  16. 16.
    Snyder, L. R., Glajch, J. L., and Kirkland, J. J. (1988) Non-ionic samples: reversed-and normal-phase HPLC. In: Practical HPLC Method Development. Wiley, New York, NY, pp. 233–291.Google Scholar
  17. 17.
    Snyder, L. R., Stadalius, M., and Quarry, M. A. (1983) Gradient elution in reversed-phase HPLC separation of macromolecules. Anal. Chem. 55, 1412A–1430A.CrossRefGoogle Scholar

Copyright information

© Humana Press Inc. 2007

Authors and Affiliations

  • Vladimir V. Tolstikov
    • 1
  • Oliver Fiehn
    • 1
  • Nobuo Tanaka
    • 2
  1. 1.Genome CenterUniversity of California-DavisDavis
  2. 2.Department of Polymer Science and EngineeringKyoto Institute of TechnologyKyotoJapan

Personalised recommendations