The Use of Dipeptide Supplementation as a Means of Mitigating the Negative Effects of Dietary Soybean Meal on Zebrafish Danio rerio

Soybean meal (SBM) is the most common source of protein used to replace fishmeal (FM) in aquaculture diets. SBM inclusion in diets has been found to negatively affect growth and induce intestinal inflammation in fish. The objective of this study was to determine the effect of health-promoting dipeptide (carnosine, anserine, alanyl-glutamine) supplementation into SBM-based feeds on growth performance, intestinal health, and muscle free amino acid composition, an indicator of dietary amino acid availability, in a zebrafish model. There were 5 treatment groups in this study. The first group ((+) Control) received a FM-based diet. The second group ((-) Control) received SBM-based diet. The last three groups (Ala-Glu, Car, and Ans) were fed SBM-based diets, supplemented with alanyl-glutamine, carnosine, and anserine respectively. All groups received their respective diets during 33-59 dph. The Ala-Glu and Car groups experienced a significantly higher weight gain than the (-) Control group, weighing 35.38% and 33.96% more, respectively at the conclusion of the study. There were no significant differences among gene expression between the groups, but Ala-Glu had the highest expression of both nutrient absorption genes measured, PepT1 and fabp2. Ala-Glu had significantly longer intestinal villi, and a significantly higher villus length-to-width ratio than the (-) Control group. Among the free amino acid composition, the Car group had a significantly higher post-prandial concentration of lysine, compared to the (-) Control group. The increase in villi surface area and expression of nutrient absorption genes represent an improvement in intestinal absorptive capacity in the Ala-Glu group. The increase in lysine concentration may signify an increase in the retention of protein in the fish in the Car group. The results from this study provide support for the use of alanyl-glutamine and carnosine supplementation as a means of improving growth performance of zebrafish fed with a 100% SBM-based diet.

160 Fish were observed each day for signs of distress. No fish in this study had to be euthanized 161 prior to the completion of the experiment.

162
All experiments were carried using a recirculated aquaculture system (Pentair Aquatic 163 Eco-systems, Cary, NC). The system consisted of 3 L and 5 L tanks, stacked in rows. Water in the 164 system was constantly filtered and recirculated using a mat filter, UV light, a carbon filter, and 165 biofiltration. The average water temperature was 27.1 0.2 o C, the pH was 7.01 0.23, and ± ± 166 the salinity was kept between 1 and 3 ppt (higher during the first feeding to prolong the 167 viability of the live food [25]. The photoperiod consisted of 14 hours of darkness and 10 hours 168 of light, with the overhead illumination from 08:00-18:00. 169 Feed preparation and formulation 170 Dry components of feeds were ground to a fine particle size (0.5-0.25 mm) using a 171 centrifugal mill (Retsch 2M 100, Haan, Germany). Once ground, components were mixed 172 (Farberware Mixer, Fairfield, CA) to achieve uniform dispersion of all ingredients within the mix.  208 The fish were fed with a restricted feeding rate of 8% of the total biomass per day.

Sampling and measuring
At the conclusion of the study, three fish from each tank were sampled and stored in a 211 10% formalin solution for histology analyses, and 3 fish in each tank were euthanized with MS-     Results are presented as means (± standard deviation). One-way ANOVA was used to 281 test the data, and an LSD test was run to test the differences between groups. Differences 282 between groups are considered significant at p values < 0.05. Statistical analysis was performed 283 using R software.

Results
285 Growth performance 286 The supplementation of alanyl-glutamine and carnosine had a significant effect on the 287 growth performance of zebrafish ( There were no significant differences in intestinal expression of il1b, il10, tnfα, mmp9, 300 fabp2, or PepT1 among the groups (Figure 1). One trend observed in the data is that both 301 nutrient absorption genes, PepT1 and fabp2 tended to be numerically higher in Ala-Glu group 302 compared to (-) Control, Car and Ans groups. The results for intestinal histology are presented in Table 4. Ala-Glu and Ans had 310 significantly longer proximal villi than the (-) Control and Car groups. There were no significant 311 differences in the widths of the intestinal villi between groups. Using the length and width 312 measurements of the intestinal villi, a length-to-width ratio for the villi was calculated to 313 represent the surface area available for nutrient absorption. The Ala-Glu group had a 314 significantly higher length-to-width ratio of proximal villi than both the (-) Control and Car 315 groups. The Ans group had a significantly higher length-to-width ratio than the Car group, and 316 the length-to-width ratio of the (+) Control group was not significantly different from any of the 317 experimental groups.
318 Table 4. Treatment effect on the villus measurements in the proximal portion of the intestine. 319 Values are presented as means (± std. dev). Superscript letters indicate statistical significance 320 between groups. The significance was determined using a One-Way ANOVA and an LSD test 321 with a p value <0.05. The results for muscle FAA postprandial levels 3-hours after feeding are presented in 325 Table 5. Out of 20 FAA analyzed, only 9 showed statistical differences among groups. Among  (Table 5).

Proximal Villus Measurements
356 Previous studies have found that high dietary inclusion of PP significantly decreases the