Abstract
Epicuticular waxes on the surface of plant leaves are important for the tolerance to abiotic stresses and plant–parasite interactions. In the onion (Allium cepa L.), the variation for the amounts and types of epicuticular waxes is significantly associated with less feeding damage by the insect Thrips tabaci (thrips). Epicuticular wax profiles are measured using used gas chromatography mass spectrometry (GCMS), which is a labor intensive and relatively expensive approach. Biochemical spectroscopy is a non-destructive tool for measurement and analysis of physiological and chemical features of plants. This study used GCMS and full-range biochemical spectroscopy to characterize epicuticular waxes on seven onion accessions with visually glossy (low wax), semi-glossy (intermediate wax), or waxy (copious wax) foliage, as well as a segregating family from the cross of glossy and waxy onions. In agreement with previous studies, GCMS revealed that the three main waxes on the leaves of a wild type waxy onion were the ketone hentriacontanone-16 (H16) and fatty alcohols octacosanol-1 (Oct) and triacontanol-1 (Tri). The glossy cultivar “Odourless Greenleaf” had a unique phenotype with essentially no H16 and Tri and higher amounts of Oct and the fatty alcohol hexacosanol-1 (Hex). Hyperspectral reflectance profiles were measured on leaves of the onion accessions and segregating family, and partial least-squares regression (PLSR) was utilized to generate a spectral coefficient for every wavelength and prediction models for the amounts of the three major wax components. PLSR predictions were robust with independent validation coefficients of determination at 0.72, 0.70, and 0.42 for H16, Oct, and Tri, respectively. The predicted amounts of H16, Oct, and Tri are the result of an additive effect of multiple spectral features of different intensities. The variation of reflectance for H16, Oct, and Tri revealed unique spectral features at 2259 nm, 645 nm, and 730 nm, respectively. Reflectance spectroscopy successfully revealed a major quantitative trait locus (QTL) for amounts of H16, Oct, and Tri in the segregating family, agreeing with previous genetic studies. This study demonstrates that hyperspectral signatures can be used for non-destructive measurement of major waxes on onion leaves as a basis for rapid plant assessment in support of developing thrips-resistant onions.
Highlights
Foliar spectroscopy uses reflectance or absorption features associated with molecular bonds to both rapidly and non-destructively estimate amounts of specific chemicals in living plant tissue [1,2,3,4].Reflectance spectroscopy, known as hyperspectral sensing, consists of continuous measurements across a broad spectral range, e.g., every 3 nm in the UV/visible to near-infrared (VNIR, 350–1000 nm) or to shortwave infrared (VSWIR, 350–2500 nm)
The logarithm of odds (LOD) score for Tri was higher with spectroscopy than gas chromatography mass spectrometry (GCMS) at 9.1 versus 5.5, respectively, and the prediction model using hyperspectral tools with the R2 cross-validation dataset was almost identical to the external validation dataset. These results demonstrate that mapping and genetic effects using predicted spectroscopic values (Table 6) detected the same major quantitative trait locus (QTL) as GCMS [34]
Reflectance spectroscopy was used for non-destructive measurement of amounts of major epicuticular waxes on onion leaves and revealed distinct absorption features in the visible (645 nm), near infrared (730 nm), and SWIR (2259 nm) for Oct, Tri, and H16, respectively
Summary
Foliar spectroscopy uses reflectance or absorption features associated with molecular bonds to both rapidly and non-destructively estimate amounts of specific chemicals in living plant tissue [1,2,3,4].Reflectance spectroscopy, known as hyperspectral sensing, consists of continuous measurements across a broad spectral range, e.g., every 3 nm in the UV/visible to near-infrared (VNIR, 350–1000 nm) or to shortwave infrared (VSWIR, 350–2500 nm). Foliar spectroscopy uses reflectance or absorption features associated with molecular bonds to both rapidly and non-destructively estimate amounts of specific chemicals in living plant tissue [1,2,3,4]. Spectroscopy has been successfully used to estimate physiological and biochemical features of plants. Spectroscopy on green leaves has been used to measure nitrogen concentration in tissues from a range of crops [9,10,11], as well as physiological parameters such as rates of ribulose-1,5-bisphosphate-carboxylation as an estimate of photosynthetic capacity [12] and other constituents such as phenolics that are important for plant defense [1,13,14,15]. Hyperspectral data have been utilized to differentiate the reflectance fingerprints of susceptible versus resistant phenotypes for the pathogen Cercospora beticola in sugar beet [16], damage by Thrips tabaci L. (thrips) on cabbage [17,18,19], and numbers of microsclerotia of the fungus Macrophomina phaseolina (Tassi) Goid on soybean [18], supporting spectroscopy as a selection tool for plant breeding
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
