This project investigated the use of new near infrared transmittance (NIT) technology for individual kernel sorting to reduce Fusarium damaged kernels (FDK) and mycotoxins in grain. There were two objectives: 1) determine the efficiency of sorting; and 2) assess how highly contaminated sorted grain fractions can be used in dietary challenges for broilers as a screen for methods to reduce mycotoxin exposure.
Fusarium damaged kernels are associated with lower crude protein (CP) caused by fungal infestation during kernel development, and may contain varying concentrations of mycotoxins (e.g. deoxynivalenol; DON). The BoMill TriQ measures the NIT of limited spectra to predict CP variation among individual kernels at ~2 - 3 MT/hour. Five sources of downgraded grain attained from grain producers in Western Canada in 2013 were sorted into ten calibration fractions, each analyzed for CP, FDK and 16 common mycotoxins. From these analyses, three wheat sources were individually sorted into three test fractions: outliers (10%); high FDK (low CP; 20% of source); and low FDK (high CP; 70% of source). Four diet recombinations were produced based on increasing inclusion of the high % FDK fraction [0% (M0), 20% (M20), 40% (M40) or 60% (M60)] of each wheat source, providing increasing mycotoxin concentrations in the test diets. Productions of these diets from re-combining the FDK fractions enabled a 3 wheat source x 4 FDK level (M0, M20, M40, M60) factorial design. The 12 test diets were included at 70 (starter, 0 - 21 d) and 75% (grow/finisher, 21 - 35 d) of a basal diet. Diets were formulated to meet or exceed NRC (1994) requirements for broilers.
Eight cages of five, one-day old male Ross 308 broilers were each randomly assigned to the 12 starter diets. The number of cages were reduced to three per diet at 21 d. Broiler performance were recorded for the 0 - 21 and 21 - 35 d. Apparent metabolizable energy (AME; kcal ME/kg diet) and nitrogen retention (NR; %) were determined using digestible markers and excreta collections. Five biomarkers of immune function were measured for starter and grower/finisher periods: 1) cell-mediated immune response to injection of the T cell mitogen phytohemagglutinin (PHA); 2) humoral response to immunization with bovine serum albumin (BSA) antigen; 3) relative weights of liver, spleen and bursa of Fabricius; 4) heterophil to lymphocyte (H:L) ratio; and 5) histopathology of primary and secondary immune organs. Analysis of sorting efficiency of this technology indicated that grain could be separated into 10% increments based on unique spectral ranges and their correlation to the chemical characteristics of CP. Indications were that the lowest 20% CP kernels contained increased FDK (15.4%) and DON (10.2 ppm) compared to the unsorted kernels (2.4% and 1.7 ppm). The statistical correlations between FDK, DON and CP provided the capability to produce high and low mycotoxin fractions for use in the poultry feeding trial. Analysis of growth and performance endpoints of each exposure period indicated no significant difference (P > 0.05), however AME and NR were different (P < 0.01) among treatment groups at 21 and 35 d. Analysis of immune system endpoints indicated no significant differences (P > 0.05) among treatment groups in cell-mediated (PHA; 0.32 - 0.35 % change), humoral (BSA; 0.57 - 0.64 % change) or H:L ratio (0.03 - 0.13 % change) immune responses. However, histopathological examination of the spleen (P < 0.05) at 21 d and the liver (P < 0.01) at 35 d showed increases in lymphoid aggregates and/or granulopoisis in the diet containing 8 ppm DON suggesting potential adverse effects on the immune system.
Overall, the results of these studies indicate that the NIT technology has the potential to produce naturally contaminated diets with various levels of mycotoxins from a single source of grain. These naturally contaminated diets may improve our ability to evaluate models to examine the effects of mycotoxin exposures to poultry or livestock.
| Identifer | oai:union.ndltd.org:USASK/oai:ecommons.usask.ca:10388/ETD-2015-08-2149 |
| Date | 2015 August 1900 |
| Contributors | Scott, Tom A., Wickstrom, Mark L. |
| Source Sets | University of Saskatchewan Library |
| Language | English |
| Detected Language | English |
| Type | text, thesis |
Page generated in 0.0021 seconds