1. Efficacy testing of a glucomannan mycotoxin binder towards the effects of feed-borne Fusarium mycotoxins in turkey poults based on specific and unspecific parameters Mathias Devreese 1,2 *, George Girgis 1, Si-Trung Tran 2, Siegrid De Baere 2, Patrick De Backer 2, Siska Croubels 2, Trevor K. Smith 1 1 Department of Animal and Poultry Science, Ontario Agricultural College, University of Guelph, 50 Stone Road East, Guelph N1G 2W1, Ontario, Canada 2 Department of Pharmacology, Toxicology and Biochemistry, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, 9820 Merelbeke, Belgium e-mail: mathias.devreese@ugent.be Introduction and Aims Mycotoxin binders are commonly added in animal feed to counteract the negative effects of mycotoxins on production animals. It is obvious that these products should be tested for their mycotoxin adsorbing ability in the gut. Discussion is ongoing, however, by which method these products should be evaluated in vivo. Mycotoxin binders are generally evaluated based on unspecific parameters such as performance data, histological changes and/or alterations in immune responses. These criteria are unspecific and do not show a direct correlation with the adsorption capacity in the intestinal tract. Other, direct end-points have recently been proposed by the European Food Safety Authority (EFSA) (EFSA, 2010). These are based on specific toxicokinetic parameters. The most relevant end-point for deoxynivalenol (DON), for example, is DON and the main metabolite de- epoxydeoxynivalenol (DOM-1) in plasma. The goal of this study was to evaluate the efficacy a glucomannan (GMA) mycotoxin binder on both sets of parameters in turkey poults. Materials and Methods Two hundred forty one-day-old turkeys were randomly allocated to one of four diets: (1) control (minimally contaminated) (2) control + 0.2 % GMA (3) naturally contaminated (4-6 mg DON/kg feed) (4) naturally contaminated + 0.2 % GMA. At the end of starter (3 w), grower (6 w), developer (9 w) and finisher (12 w) phases, plasma concentrations of DON and DOM-1, body weight gain, feed intake, feed conversion rate and plasma biochemistry were evaluated. Plasma biochemistry profile included evaluation of following components: Ca, P, total protein, albumin, globulin, albumin:globulin ratio, glucose, cholesterol, total bilirubin, γ-glutamyltransferase (GGT), aspartate aminotransferase (AST), creatine kinase (CK), amylase, lipase, uric acid, lactate dehydrogenase (LDH), bile acid and glutamate dehydrogenase (GLDH). At the end of the starter phase, duodenal sections were excised from 4 birds/pen (12/diet) for histological (morphometry) and immunohistochemistry analysis (CD8 + T-lymphocyte counts). Results and Discussion Acknowledgments The authors would like to thank the Agency for Innovation by Science and Technology (IWT,, the FWO-Vlaanderen, Alltech Inc. (Lexington, KY, USA) and the Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA, Canada) for their financial support. Experimental diets: The major contaminant was DON (4.0-6.5 mg/kg). Diets also included lesser amounts of 15-aDON (0.45-0.57 µg/kg), OTA, ZON, aflatoxins and fumonisins. Bird performance: Except for the starter phase, no significant differences in body weight, weight gain, feed intake or feed conversion rate (FCR) were observed. Remarkably, birds receiving the contaminated diet showed a significantly higher body weight and body weight gain in the starter phase. Plasma biochemistry: Differences were observed amongst the different dietary groups. However, no parameter was consistently altered in any rearing phase (Table 1). Duodenal morphometry: Birds fed the contaminated diet showed a significantly lower villus height (VH) and apparent villus surface area (AVSA). This negative effect was prevented by addition of GMA into the contaminated diet (Table 2). Duodenal immunohistochemistry: Higher CD8 + T-lymphocytes count was observed turkeys fed either the contaminated or the contaminated + GMA diet (Table 2). DON and DOM-1 plasma concentrations: No DON or DOM-1 was detected in the control groups. DON and DOM-1 plasma concentrations were detectable in all analyzed samples of birds fed the contaminated or the contaminated + GMA diet. No significant differences, however, were observed (Table 3). ControlControl + GMAContaminatedContaminated + GMA Starter CK (U/L) 2584.00 ± 160.407 a 3681.17 ± 232.031 b 4229.08 ± 337.814 b 4136.33 ± 62.322 b Amylase (U/L) 654.33 ± 33.422 a 663.75 ± 5.774 a,b 757.17 ± 52.218 b 705.67 ± 50.069 b Grower P (mmol/L) 2.53 ± 0.053 a 2.60 ± 0.063 a,b 2.43 ± 0.061 a 2.93 ± 0.055 b Developer LDH (U/L) 488.33 ± 21.586 a 597.33 ± 16.807 b 576.33 ± 5.132 a,b 510.33 ± 7.441 a,b Finisher Cholesterol (mmol/L) 3.52 ± 0.073 a,b 3.23 ± 0.032 a 3.28 ± 0.058 a 3.74 ± 0.042 b A different superscript letter indicates a significant difference (P < 0.05) Values represent the overall mean of the replicate means (n=3) ± SEM ControlControl + GMAContaminatedContaminated + GMA Morphometry Villus height (µm) 2205.12 ± 35.4812222.45 ± 30.2091952.25 ± 23.974 ***2253.29 ± 21.949 Crypt depth (µm) 239.59 ± 17.750250.96 ± 6.346243.60 ± 9.846252.37 ± 7.486 Submucosa thickness (µm) 29.61 ± 0.33628.41 ± 0.65927.58 ± 0.50127.68 ± 0.385 Villus width down (µm) 134.87 ± 2.452137.75 ± 1.704140.35 ± 2.148132.60 ± 1.175 Villus width up (µm) 134.79 ± 1.208139.47 ± 1.456136.63 ± 1.221133.79 ± 1.012 Mean villus width (µm) 134.83 ± 1.568138.61 ± 1.446138.49 ± 1.499133.19 ± 1.015 Apparent villus surface area (cm²) 2977.03 ± 689.763077.88 ± 516.972701.71 ± 51.642 *3004.66 ± 450.27 CD8 + T-lymphocytes Positivity (%) 2.53 ± 0.2932.77 ± 0.3844.89 ± 0.391 *6.04 ± 0.512 *** * and *** indicate a significant difference compared to the control at 0.05 < P < 0.01 and P <0.001 respectively Values represent the overall mean of the replicate means (n=12) ± SEM ControlControl + GMAContaminatedContaminated + GMA Week 1 DONND 1.14 ± 0.3341.27 ± 0.337 DOM-1ND 2.45 ± 0.1982.82 ± 0.178 Starter Diet DONND 2.09 ± 0.4902.28 ± 0.415 DOM-1ND 2.28 ± 0.0432.40 ± 0.048 Grower Diet DONND 3.21 ± 0.3373.06 ± 0.409 DOM-1ND 9.51 ± 0.6389.12 ± 1.230 Developer Diet DONND 1.71 ± 0.2121.38 ± 0.267 DOM-1ND 4.00 ± 0.2103.73 ± 0.419 Finisher Diet DONND 1.00 ± 0.0361.11 ± 0.133 DOM-1ND 2.11 ± 0.2172.20 ± 0.352 ND = Not Detected (<LOD) Values represent the overall mean of the replicate means (n=3) ± SEM Table 1. Altered plasma biochemistry parameters in turkey poults fed the experimental dietsTable 2. Morphometrical and immunohistochemical analysis of the duodenum at the end of the starter phase (3 weeks) Table 3. Plasma concentrations (ng/mL) of deoxynivalenol (DON) and de-epoxydeoxynivalenol (DOM-1) after feeding different experimental diets Feeding naturally DON contaminated diets to turkey poults altered some unspecific parameters such as growth rate, plasma biochemistry profile, duodenal villus height and apparent villus surface area and CD8 + T-lymphocyte count in the duodenum. A yeast derived mycotoxin binder, GMA, was partially effective in preventing those effects. Performance parameters and plasma biochemistry profiles were not found suitable to evaluate the efficacy of mycotoxin binders on DON absorption in turkey poults as they were not consistent. GMA was able to counteract the negative effects of DON on duodenal morphometry but did not alter the increased influx of CD8 + T-lymphocytes. Plasma concentrations of DON and DOM-1 were not altered by the addition of GMA to the diet, suggesting that GMA was ineffective in decreasing DON absorption. These data suggest that the beneficial effects of GMA are due to another mechanism than DON adsorption in the gut. www.mytox.be