In-field volatile collection and analysis method utilizing a portable GC-MS: in situ headspace analysis of intact and damaged yellow starthistle flower heads

Authors: Beck, John; PORTER, NATHAN - Torion Technologies, Inc; Cook, Daniel; GRIFFITH, COREY - Former ARS Employee; Gee, Wai; RANDS, ANTHONY - Torion Technologies, Inc; TRUONG, TAI - Torion Technologies, Inc; Smith, Lincoln; SAN ROMAN, ITXASO - University Of Basque Country

Submitted to: Phytochemical Analysis
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/29/2015
Publication Date: 6/11/2015
Publication URL: http://handle.nal.usda.gov/10113/61633
Citation: Beck, J.J., Porter, N., Cook, D., Griffith, C., Gee, W.S., Rands, A.D., Truong, T.V., Smith, L., San Roman, I. 2015. In-field volatile collection and analysis method utilizing a portable GC-MS: in situ headspace analysis of intact and damaged yellow starthistle flower heads. Phytochemical Analysis. 26:395-403.

Interpretive Summary: Understanding the complex odors that form the basis of chemical communication of plants and insects is an important component of chemical ecology. Accordingly, the collection of these chemical cues from plants in their normal environment is integral to the understanding of multifaceted plant-insect communications. Often times remote plant locations and the need for a large number of replicates can make in situ (on-plant) odor analyses a daunting logistical challenge. A portable mass spectrometer detector-based portable gas chromatograph system utilizing needle trap absorbent technology for dynamic volatile collection was used to discriminate between damaged and intact Centaurea solstitialis (yellow starthistle) flower heads in both a natural and potted-plant setting. Yellow starthistle is considered an invasive weed of the western US. Odors collected onto the dynamic needle trap assembly were directly desorbed onto a portable GC-MS system, which detected a total of 36 odors from the four treatments. Statistical analysis showed four distinct clusterings of data representing the four treatments – intact and damaged potted plant, and intact and damaged natural plant. Additional statistical analysis provided specific odors that may serve as signals from the treatments, among them the compound cyclosativene that signified damaged plant tissue.

Technical Abstract: Introduction: Understanding the complex chemical signaling of plants and insects is an important component of chemical ecology. Accordingly, the collection of chemical cues from plants in their normal environment is integral to elucidation of multifaceted plant-insect communications. Often times remote plant locations and the need for a large number of replicates can make in situ headspace analyses a daunting logistical challenge. A portable mass spectrometer detector-based portable gas chromatograph system utilizing needle trap absorbent technology for dynamic volatile collection was used to discriminate between damaged and intact Centaurea solstitialis (yellow starthistle) flower heads in both a natural and potted-plant setting. Yellow starthistle is an invasive weed of the western US. Objective: Using a unique needle trap dynamic volatile collection and desorption system, determine if a portable GC-MS system was capable of distinguishing between intact and mechanically damaged plant treatments, and differences in plant locale/environments. Methodology: A portable mass spectrometer detector-based portable gas chromatograph system utilizing needle trap adsorbent technology for dynamic volatile collection was used to collect and analyze the in situ headspace volatiles of varying yellow starthistle treatments. Discriminant analysis was used to distinguish treatments and identify biomarker volatiles. Analysis of variance (ANOVA) was used to determine significant differences between treatment volatile amounts. Results: Volatiles collected onto a polymer medium in a dynamic needle trap assembly were directly desorbed onto a portable GC-MS system. The portable GC-MS system detected a total of 36 volatiles over the four treatments. Each GC-MS run was completed in less than three minutes. Discriminant analysis showed four distinct clusters representing the four treatments – intact and damaged potted plant, and intact and damaged natural plant. ANOVA and pairwise testing of peak areas identified specific volatiles that demonstrated significant differences between treatments. Conclusion: Discriminant analyses of data from 6-8 replicates for each treatment provided four clusters demonstrating the ability of the portable system to distinguish the treatments based on their detected volatile profiles. Additional statistical analysis provided specific biomarker volatiles for the treatments, among them cyclosativene signifying damaged plant tissue.