Characterization of autoclaved flaxseed as feed for ruminants using conventional and mid-IR spectroscopic based approaches

• Main Author: Doiron, Kevin
• Other Authors: Tyler, Robert T. (Bob)
• Format: Others
• Language: en
• Published: University of Saskatchewan 2009
• Subjects: secondary structure; oeb; dve; protein; cncps; s-ftir; drift; flaxseed
• Online Access: http://library.usask.ca/theses/available/etd-03252009-141827/

The objectives of this study were to investigate the effects of autoclave heating on the rumen protein degradation characteristics of flaxseed (<i>Linum usitatissimum</i>, cv. Vimy), and to compare them to differences in diffuse reflectance infrared Fourier transform (DRIFT) and Synchrotron based Fourier transform infrared microspectroscopy (S-FTIR) measurements of the protein alpha-helix to beta-sheet ratios. Hierarchical cluster analysis (CLA) and principal components analysis (PCA) were also conducted to identify differences in the DRIFT spectra. Flaxseed samples were kept raw for control or autoclaved in batches at 120°C for 20, 40 or 60 min for treatments 1, 2 and 3, respectively. The rumen degradation kinetics of protein were measured along with the protein sub-fractions of the Cornell net carbohydrate and protein system (CNCPS), and chemical composition. Intestinal digestibility was determined using the three-step procedure outlined by Calsamiglia and Stern (1995). Protein supply to the small intestine was determined using the NRC (2001) and DVE/OEB models. The results showed that heating increased dry matter (DM) and ether extract (EE) content, while reducing neutral detergent fibre (NDF) and acid detergent fibre (ADF), with little numerical difference between the three treatments. Soluble crude protein (SCP) also decreased upon autoclaving with concomitant increases in non-protein nitrogen (NPN), neutral detergent insoluble nitrogen (NDIN) and acid detergent insoluble nitrogen (ADIN). The CNCPS protein sub-fractions with the greatest changes were the buffer-soluble true protein fraction (PB1) and the fraction representing buffer-insoluble true protein which is not bound to NDF (PB2) showing dramatic increases, indicating a decrease in the overall protein degradability. <i>In situ</i>experiments showed a reduction in effective degradable dry matter (EDDM) as well as a reduction in effective degradable crude protein (EDCP) without significant differences between the treatments. Intestinal digestibility of protein as estimated by the three-step procedure showed no changes upon autoclaving. Modeling results, with flaxseed as the only feed source, for absorbable ruminally-undegraded feed protein in the intestines using both the NRC (2001) and DVE/OEB systems showed increases as a consequence of the autoclave treatments but again there were no differences between the treatments. The degraded protein balance results showed for both the NRC (2001) and DVE/OEB models that both were decreased upon autoclave treatment. However, the values for the NRC (2001) model suggested a potential nitrogen (N) deficiency and, therefore potentially impaired microbial crude protein (MCP) production, whereas the values for the DVE/OEB system showed potential N excess and, therefore, possible loss from the rumen. DRIFT analysis of protein secondary structure ratios showed a decrease in the alpha-helix to beta-sheet ratio for the whole seed, whereas results from S-FTIR spot data for cotyledon tissue showed autoclaving had the opposite effect on the ratio. CLA and PCA were successfully used to make distinctions between the different treatment spectra and showed enhanced sensitivity upon selection of a smaller spectral window to include only the amide I and II portion of the IR spectrum. The results failed to demonstrate any differences between the autoclave treatments used in this study, and showed that autoclaving generally decreased effectively ruminal degradability of flaxseed protein. The results further indicated that autoclaving had a significant enough effect on the flaxseed to permit identification of the altered alpha-helix to beta-sheet ratio with the mid-IR spectrum, as well as differentiation between the treatments using PCA and CLA. PCA and CLA results suggest that mid-IR spectral methods are more sensitive than traditional methods when used to identify differences between the heat treatments.