Fourier transform infrared spectroscopy (FTIR) is a technique used for probing chemical structure of molecules and their complexities. In FTIR spectrum, infrared absorption at specific wavenumber corresponds to the energy of the vibration of functional groups in the samples, which offers information on molecular structures and their compositions. FTIR has an ability to examine supramolecular structures without destructive and complicated sample preparation requirements. Besides, it is compatible with remote sensing in the fields. In the current study, FTIR was used for probing metabolomic behavior of wheat under heat stress.

To establish a protocol for FTIR-based metabolome profiling, FTIR spectroscopy was combined with chemometrics as below. Wheat genotype “Norin 61” was exposed to heat stress (42oC) for three days at three leaves stage, and subjected to FTIR spectroscopy. Principal component analysis of the spectral data showed an obvious overlapping feature between control and heat-stressed plants. Thus, supervised machine learning through linear discriminant analysis (LDA) was applied and clear discrimination between heat and control leaves was observed. Major LDA loadings were in the fingerprint region (400-1800 cm−1). Wavenumbers with the high LDA loadings were chosen for the construction of spectrum-based biomarkers, which distinguished heat-stressed leaves from the controls. Six markers were developed; two showed an increase under heat stress (1729 cm-1 and 1465 cm-1), while the other four showed a decrease under the stress (1502 cm-1, 1251 cm-1, 576 cm-1, and 482 cm-1). Among these markers, the wavenumber 1729 cm-1 region can be assigned to the stretching vibrations of ester C=O groups, which are rich in the polar interfacial regions of membrane lipids, suggesting a presence of heat tolerance mechanism for membrane stabilization in heat stressed leaves. The wavenumber 1502 cm-1 can be potentially annotated as the aromatic skeletal vibration of lignin, suggesting a possibility that the structure and/or composition of cell wall components were altered under heat stress in wheat leaves. Overall, these findings demonstrated the utility of FTIR-based chemical fingerprints for profiling of wheat under heat stress.