Location: Plant Science ResearchTitle: NIRS Prediction of Corn Stover Cell Wall Composition and Conversion Potential, and Relationships among these Traits Author
|Jung, Hans Joachim|
Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 1/5/2009
Publication Date: 2/9/2009
Citation: Jung, H.G., Lewis, M.L., Lorenzana, R.E., Bernardo, R. 2009. NIRS Prediction of Corn Stover Cell Wall Composition and Conversion Potential, and Relationships among these Traits [abstract]. Genomics: GTL Awardee Workshop VII and USDA-DOE Plant Feedstock Genomics for Bioenergy Awardee Workshop, February 8-11, 2009, Bethesda, Maryland. p. 53-55. Interpretive Summary:
Technical Abstract: Implementation of a breeding program for corn stover quality traits related to cellulosic ethanol production is dependent on having a rapid and inexpensive system for phenotyping plant material. We report here on the near-infrared reflectance (NIRS) prediction equations we have developed for phenotyping corn stover. Testcrosses of 223 recombinant inbred lines (RIL) derived from B73 x Mo17 and the two parental inbreds were grown at four locations in Minnesota in 2007. Corn stover, minus the ear (grain, cob, husk, and shank), was collected at grain maturity. Stover was chopped, dried at 60 degrees C, and ground prior to scanning by NIRS (1100 to 2500 nm). A calibration sample set (N=154) was analyzed for cell wall composition by a two-stage acid hydrolysis method and conversion potential in a dilute acid/high temperature pretreatment and enzymatic saccharification process. Percent cell wall sugars released by the conversion process were determined as the difference between stover composition and the wall sugars remaining in the insoluble residue from the conversion process. NIRS calibration equations for stover quality were developed by modified partial least squares regression. The NIRS calibration equation statistics indicated that stover cell wall components can be predicted with varying precision. Glucose (primarily from cellulose) was well predicted but the other major components gave only moderate fits in the calibrations. Better calibrations have been observed for cell wall composition of corn plant part, alfalfa stem, and switchgrass herbage sample sets; however, those studies included more variability in composition due to multiple maturity stages. It was observed that calibration of glucose release by the conversion process had a high R2 whereas release of other cell wall sugars gave poor calibrations. In contrast, in vitro rumen digestibility of all cell wall sugars have high calibration statistics. This difference can be explained by the fact that glucose release in the conversion process and rumen digestibility are both enzymatic processes and cell wall matrix structure interacts with enzyme accessibility to polysaccharides. This is unlike hydrolysis of non-glucan polysaccharides in the conversion process which was primarily an acid reaction that should be less influenced by wall matrix structure.