Comprehensive Two-Dimensional Gas Chromatography (GCxGC)—Time-of-Flight Mass Spectrometry of Metabolic Samples Obtained from the Leaves of a Hybrid Aspen (Populus Tremula) Plant

Significance of the topic

The comprehensive characterization of small-molecule metabolites in plants is central to functional genomics, chemical ecology, breeding, and environmental physiology. GC-based metabolomics remains a core platform for volatile and derivatizable primary metabolites because it delivers high sensitivity, reproducibility and well‑established spectral libraries. However, high sample complexity and coelution limit one‑dimensional GC approaches. Combining comprehensive two‑dimensional gas chromatography (GCxGC) with fast time‑of‑flight mass spectrometry (TOFMS) and automated spectral deconvolution materially increases analytical peak capacity, enabling deeper coverage of the volatilome/derivatizable metabolome in short run times.

Objectives and study overview

This study aimed to evaluate GCxGC‑TOFMS for rapid metabolic profiling of hybrid aspen (Populus tremula) leaves. Dried extracts from two different leaves (positions 9 and 20) of five individual plants were derivatized and analyzed to (i) maximize the number of detectable metabolites in a single ~20‑minute GCxGC run, (ii) demonstrate automated processing and sample comparison, and (iii) illustrate the benefit of spectral deconvolution for resolving severe co‑elution.

Methodology

Sample preparation:

• Plants grown under long‑day conditions for 15 weeks; leaves numbered from the top (leaf 9 and leaf 20) taken from five plants.
• Dried extracts were spiked with nine isotopically labeled internal standards prior to derivatization.
• Derivatization: methoxyamination with methoxyamine hydrochloride (15 mg/mL in pyridine), 30 µL, heated 60°C for 30 min then 16 h at room temperature; silylation with 30 µL MSTFA + 1% TMCS for 1 h at room temperature. Analysis performed within 12 h of derivatization.

Chromatography and acquisition conditions (summary):

• Instrument platform: Agilent 6890 GC fitted with a LECO thermal modulator (Zoex technology licensed) coupled to a LECO Pegasus TOFMS.
• Primary column: Rtx‑5, 10 m × 0.18 mm i.d., 0.18 µm film; secondary column: DB‑17, 1 m × 0.1 mm i.d., 0.1 µm film.
• Oven program: 70°C (2 min) to 320°C (5 min) at 20°C/min; secondary oven set 5°C offset from main oven.
• Modulation: 3.0 s period with 0.5 s hot pulse; modulator set 30°C offset from main oven.
• Injection: 1 µL split 10:1, inlet 270°C, He carrier at 1.0 mL/min.
• MS: EI 70 eV; mass range 60–800 u; acquisition 150 spectra/s; source 200°C.

Data processing:

• ChromaTOF software used for automated peak finding, spectral deconvolution, library matching and inter‑sample comparison (compare algorithm and reference table workflow).
• Reference sample (L9 1) processed at S/N = 100 to build a reference table of peaks; subsequent samples compared against this reference; one‑point calibration used for relative quantitation.

Used instrumentation

All instrumentation explicitly reported in the study:

• Agilent 6890 gas chromatograph with LECO thermal modulator (Zoex license)
• Columns: Rtx‑5 (10 m, 0.18 mm i.d., 0.18 µm) and DB‑17 (1 m, 0.1 mm i.d., 0.1 µm)
• LECO Pegasus TOFMS (EI 70 eV)
• Data system: ChromaTOF software (LECO) for deconvolution, library search and sample comparison

Main results and discussion

Peak detection and coverage:

• The reference sample (L9 1) yielded 690 automatically detected features. After excluding modulation artifacts (column bleed, solvent tailing), more than 400 unique analytes were considered real leaf metabolites — roughly double the number typically found by comparable one‑dimensional GC‑TOFMS under similar conditions (≈200 peaks at S/N = 10).

Library matching and identification confidence:

• Identifications were primarily tentative, relying on first NIST library hits for trimethylsilyl derivatives. For some compounds individual standards were run to confirm identity (e.g., citric acid TMS derivative matched reference spectrum at 965/1000).
• Many silylated plant metabolites are absent or underrepresented in commercial libraries, so spectral matches should be interpreted cautiously.

Spectral deconvolution and separation gains:

• GCxGC provided orthogonal retention that resolved multiple analytes that co‑elute in the first dimension. Projection of 2nd‑dimension retention times demonstrated separation of components that would be indistinguishable with a single column.
• Automated deconvolution reliably extracted pure component spectra even in severe co‑elution (example: two completely co‑eluting peaks where deconvolution identified the citric acid tetramethylsilyl derivative with high confidence).

Comparative analysis across samples:

• A 40‑peak reference table was created from three chromatographic regions of interest. Relative quantitation (reference sample set to 100%) revealed biological variability across plants: ~30% of monitored analytes increased from plant 1 to plant 5, ~30% decreased, ~20% remained approximately constant, and a minority showed inconsistent trends.
• Differences between leaves 9 and 20 from the same plant were often larger than between the same leaf position across plants, indicating positional or developmental metabolite variation.

Throughput and practical runtime:

• All analyses were completed in under 20 minutes per GCxGC run, demonstrating high throughput compatible with metabolomics screening.

Benefits and practical applications

Key advantages demonstrated by this work:

• Substantially deeper coverage of derivatizable plant metabolites in a single, short chromatographic run compared with 1D GC‑TOFMS.
• Improved resolution for complex matrices via orthogonal GCxGC separations and additional effective separation by spectral deconvolution.
• Automated workflows (ChromaTOF compare/reference‑table) facilitate relative quantitation and inter‑sample comparison for fingerprinting, genotype/ecotype discrimination, and response studies (e.g., stress, treatment).

Typical applications include plant metabolite fingerprinting, screening of mutants or transgenic lines vs wild type, and studies relating environmental or experimental stimuli to pathway perturbations.

Future trends and potential uses

Anticipated directions and opportunities to extend this work:

• Expansion and curation of spectral libraries for silylated plant metabolites to improve identification confidence.
• Integration of high‑resolution MS (HRTOF, Orbitrap) or complementary ionization strategies to aid formula assignment and annotation of unknowns.
• Advanced chemometrics and machine‑learning models for automated feature annotation, pattern recognition, and biomarker discovery from GCxGC‑TOFMS datasets.
• Hybrid workflows combining GCxGC for volatile/derivatizable fraction and LC‑MS for polar/nonvolatile metabolites to approach more complete metabolome coverage.
• Standardization of derivatization, modulation and data‑processing parameters to enhance reproducibility in multi‑laboratory studies.

Conclusion

This study demonstrates that GCxGC‑TOFMS, when combined with robust derivatization, isotopic internal standards, automated spectral deconvolution and software‑enabled sample comparison, substantially increases the number of detectable metabolites in plant leaf extracts within a high‑throughput (~20 min) framework. The approach resolves co‑eluting species that defeat 1D separation and yields richer datasets for comparative metabolomics, albeit with the caveat that many silylated plant metabolites require better library coverage for confident identification.

References

Jonsson P.; Gullberg J.; Nordstrom A.; Kusano M.; Kowalczyk M.; Sjostrom M.; Moritz T.; Anal. Chem.; 2004; ASAP Article
LECO Corporation application note: Comprehensive Two‑Dimensional Gas Chromatography (GCxGC)—Time‑of‑Flight Mass Spectrometry of Metabolic Samples Obtained from the Leaves of a Hybrid Aspen (Populus tremula). LECO Corporation, 2008.