Konferansebidrag og faglig presentasjon » Vitenskapelig foredrag
Preliminary results on nutritional and feed technological properties of different soybean genotypes detected by near infrared spectroscopy and proteomics
XI International Feed Technological symposium.; Vrnjaka Banja. Serbia and Montenegro, 2005-05-30–2005-06-03
Hollung, Kristin; Øverland, Margareth; HRUSTIC, MILICA; SEKULIC, PETAR; MILADINOVIC, JEGOR; Martens, Harald; Narum, Bjørg; Nilsen, Bjørg Narum; Sahlstrøm, Stefan; Sørensen, Mette; Storebakken, Trond; Skrede, Anders
A total of 832 genotypes of soybeans were screened by near infrared (NIR) spectroscopy. Out of these, 30 genotypes were identified based on variation in protein content and agronomic value. Eleven genotypes were further selected to form a response surface based on high-resolution NIR spectroscopy and differences in the spectra indicative of protein, fat and carbohydrates. The samples were analysed for main carbohydrate composition (mono-, di- and oligosaccharides and non-starch polysaccharides (NSP)). Soluble proteins were separated by 2-dimensional gel electrophoresis (2DE) and protein spots were identified by matrix-assisted laser desorption/ionization time-of flight (MALDI-TOF) mass spectrometry (MS). Results indicate that there is a phenotypical variation in the chemical composition of the 30 genotypes demonstrated by high-resolution NIR spectroscopy. This information was used to select 12 samples for further investigation of carbohydrate and protein composition. Using multivariate data analysis and partial least square regression (PLSR) it was possible to study the correlations between the analysed carbohydrates and the chemical composition measured in the first NIR screening. A PLSR model showed positive correlation between total NSP content and carbohydrate content measured in the NIR screening, suggesting that high-resolution NIR is a suitable and rapid non-destructive method to determine carbohydrate composition in soybeans. In addition, we observed a positive correlation between maltose and glucose, total uronic acid and sucrose, raffinose and maltose, and raffinose and glucose. The 2DE analyses showed varying intensities of several proteins. One of the variable proteins was identified as glycinin G1 precursor. PLSR analysis using carbohydrate composition as x-variable and glycinin G1 precursor content as y-variable showed a significant negative correlation between this protein and insoluble NSP and total uronic acid. The results are being followed up by up-scaling of production and studies of feed technological and nutritional characteristics of selected genotypes.