A seven-month feeding trial with geneticallymodified (GM) Roundup Ready® (RRS®) soybeans was conducted on Atlantic salmon (initial weight 40 g) going through the parr–smolt transformation. The maternal near-isogenic soybean line was used as a non-modified control (non-GM), and the two diets were compositionally similar in all analysed nutrients. The performance and health of the fish were assessed by growth, body composition, organ development, haematological parameters, clinical plasma chemistry and lysozyme levels, with samples collected both in the freshwater- and seawater stages. Intestinal indices exhibited some differences between the groups, with the mid-intestine being consistently smaller in the GM fed fish throughout the experiment, while the distal intestine was different at one sampling point, shortly after seawater transfer. Plasma triacylglycerol (TAG) levels were higher in the GM group overall in the experiment, although the magnitude of the difference was larger around the time of seawater transfer compared to later samplings. Despite differences at individual sampling points, there were no differences in total growth during the trial. All other measured parameters showed no diet related differences. Seawater transfer caused changes in gill Na+K+-ATPase activity and plasma chloride ion concentration, as well as in haematological parameters (red blood cell count, RBC, haematocrit, Hct, haemoglobin, Hb) and plasma glucose concentration. However, both diet groups responded similarly regarding these parameters. Our overall conclusion is that the observed effects of feeding Atlantic salmon with GM soy at a 25% inclusion level were minor, and lack of consistency with previous studies suggests that they might be caused by variations in the soy strains rather than the genetic modification per se.
