Detection of Erwinia amylovora in pear leaves using a combined approach by hyperspectral reflectance and nuclear magnetic resonance spectroscopy

Erwinia amylovora infections on pear leaves were studied using nuclear magnetic resonance and hyperspectral reflectance spectroscopies. Inoculated pear plants under controlled conditions were used for comparing Erwinia amylovora infected leaves with those infected by Pseudomonas syringae pv. syringae or non-infected controls. Hyperspectral reflectance-NMR covariance analysis allowed the transfer of metabolome information obtained by NMR to hyperspectral reflectance bands through a knowledge transfer approach. At 20 d after Erwinia amylovora inoculation, correlation was found between the NMR signal at 1.16 ppm (attributed to the methyl group of a fucosyl-containing polysaccharide, identified as a specific metabolite from Erwinia amylovora) and a hyperspectral reflectance band centred at 1400 nm. At 50 d after inoculation the same marker metabolite was correlated to hyperspectral reflectance bands centred at 850 nm and 1050 nm. These methods allow maps to be developed which represent the specific infection status of pear plants, and could facilitate development of simple, fast and affordable hyperspectral reflectance-based devices for the detection of Erwinia amylovora infections on pear leaves.