Bt-maize (MON810) and non-GM soybean meal in diets for Atlantic salmon (Salmo salar L.) juveniles--impact on survival, growth performance, development, digestive function, and transcriptional expression of intestinal immune and stress responses

Abstract

Responses in Atlantic salmon (Salmo salar L.) juveniles (fry) fed diets containing genetically modified maize (Bt-maize, MON810) expressing Cry1Ab protein from first-feeding were investigated during a 99-day feeding trial. Four experimental diets were made; each diet contained ∼20% maize, either Bt-maize or its near-isogenic maternal line (non-GM maize). One pair was fishmeal-based while the other pair included standard (extracted) soybean meal (SBM; 16.7% inclusion level), with the intention of investigating responses to the maize varieties in healthy fish as well as in immunologically challenged fish with SBM-induced distal intestinal inflammation, respectively. Three replicate tanks of fry (0.17±0.01 g; initial mean weight ± SEM) were fed one of the four diets and samples were taken on days 15, 36, 48 and 99. Survival, growth performance, whole body composition, digestive function, morphology of intestine, liver and skeleton, and mRNA expression of some immune and stress response parameters in the distal intestine were evaluated. After 99 days of feeding, survival was enhanced and the intended SBM-induced inflammatory response in the distal intestine of the two groups of SBM-fed fish was absent, indicating that the juvenile salmon were tolerant to SBM. Mortality, growth performance and body composition were similar in fish fed the two maize varieties. The Bt-maize fed fish, however, displayed minor but significantly decreased digestive enzyme activities of leucine aminopeptidase and maltase, as well as decreased concentration of gut bile salts, but significantly increased amylase activity at some sampling points. Histomorphological, radiographic and mRNA expression evaluations did not reveal any biologically relevant effects of Bt-maize in the gastrointestinal tract, liver or skeleton. The results suggest that the Cry1Ab protein or other compositional differences in GM Bt-maize may cause minor alterations in intestinal responses in juvenile salmon, but without affecting overall survival, growth performance, development or health.

Figure 1. Mean activities of leucine aminopeptidase (LAP) in mmol substrate hydrolysed per hour per kg body weight (A, B and C) and maltase in µmol substrate hydrolysed per min per kg body weight (D, E and F) in whole fish and intestinal sections of Atlantic salmon juveniles fed non-GM maize or GM maize (Bt-maize) without or with soybean meal (SBM) from the first-feeding to day 99.

Means ± pooled standard errors were calculated from pooled samples of 10–15 fish per tank, three replicate tanks per treatment group (n = 3; for more details see Materials and methods section Digestive enzyme activities and bile acid concentrations). Significant GM effect (p<0.05) is indicated over the column and the trends (p<0.10) are marked with #. Significant SBM effect and GM-SBM interaction (p<0.05) are indicated by differing upper and lower case letters, respectively.

Figure 2. Mean activities of trypsin (A and B) and amylase (C and D) as well as bile salt concentration (E and F) in whole fish (WF) and intestinal sections of Atlantic salmon juveniles fed non-GM maize or GM maize (Bt-maize) without or with soybean meal (SBM) from the first-feeding to day 99.

Means ± pooled standard errors were calculated from pooled samples of 10–15 fish per tank, three replicate tanks per treatment group (n = 3; for more details see Materials and methods section Digestive enzyme activities and bile acid concentrations). Significant GM effect (p<0.05) is indicated over the columns and the trends (p<0.10) are marked with #. Significant SBM effect and GM-SBM interaction (p<0.05) are indicated by differing upper and lower case letters, respectively.

Figure 3. Representative histological detail of liver (A), pyloric caeca (B) and distal intestine (C) in Atlantic salmon juvenile fed GM maize (Bt-maize) with soybean meal (SBM) for 99 days.

No consistent differences due to dietary treatment were observed. Normal morphology of liver (A) with high level of glycogen deposits (black thick arrow) in hepatocytes; asterisk and white thick arrow illustrating portal vein and bile duct, respectively. Normal morphology of pyloric caeca (B) with distinct mucosal folds (mf), which are comprised of a single layer of normal enterocytes (e) and scattered goblet cells (arrow); the thin lamina propria (triangle) are lined with a single layer of loose connective tissue. Normal morphology of distal intestine (C) with distinct mucosal folds (mf), which are comprised of single layer of highly vacuolated enterocytes (V); enterocyte nuclei are basally located within the cells; the thin lamina propria (triangle) are lined with a single layer of loose connective tissue.

Figure 4. Radiographic examination of the skeleton in Atlantic salmon juveniles fed non-GM maize or GM maize (Bt-maize) without or with soybean meal (SBM) for 99 days.

The three figures illustrate various deformities observed in five of the 656 fish examined: vertebral fusion (arrow; A) and compressed vertebrae in neck region (oval; B and C). A normal vertebral axis is demonstrated in A and B, while C shows a slight axis deviation. See text for more details.

Figure 5. Relative mRNA expression of CD4 (A), interleukin 17a (IL17a) (B), interferon γ (IFNγ) (C), proliferating cell nuclear antigen (PCNA) (D), and heat shock protein 70 (HSP70) (E) in distal intestine of Atlantic salmon juveniles fed the non-GM maize or GM maize (Bt-maize) without or with soybean meal (SBM) for 99 days.

Means ± pooled standard errors were calculated from the means of three replicate tanks per treatment group (n = 3), with measurements performed on three fish per tank (for more details see Materials and methods section Quantitative real-time PCR [qPCR]). The significant GM effect (p = 0.0246) on IL17a expression (B) is indicated over the columns.