nd fork length were measured for 10 fish per tank. Depending on the size of fish, 220 whole fish and/ or dissected intestinal sections were pooled for 3006665 assays of digestive enzyme, mRNA expression and whole body chemical analysis. When the fish were large enough to be dissected, i.e. at 48 days of dietary exposure, different regions of the intestinal tract proximal, mid and distal intestine were collected as described earlier. Samples for histological examination were fixed in 4% buffered formaldehyde solution for 24h and subsequently stored in Digestive enzyme activities and bile acid concentrations One pooled sample of 15 whole fish per tank on days 15 and 36, and 10 whole fish per tank on days 48 and 99, respectively, were analysed for activities of pancreatic enzymes trypsin and amylase, brush border membrane enzymes leucine aminopeptidase and maltase, as well as bile acid concentration. At the last two samplings, one pooled sample of 10 dissected intestinal sections Effects of GM Bt-Maize in Diets for Juvenile Atlantic Salmon per tank was also analysed for the abovementioned parameters. On the day 48 sampling, the small size of the intestines made distinguishing the MI from the DI intestinal regions difficult. Thus the MI and DI were analysed together, while the PI, with its visible pyloric caeca, was analysed separately. On day 99, the DI was analysed separately, while MI and PI samples were analysed together. LAP and 6-Methoxy-2-benzoxazolinone chemical information Maltase activities, investigated to assess intestinal function, were analyzed following the method described previously. The frozen samples were thawed and homogenized in ice-cold 2 mM Tris/50 mM mannitol, pH 7.1, containing serine protease inhibitor Pefabloc SC. The LAP activity was measured colorimetrically using L-leucine b-naphthylamide hydrochloride as the substrate. LAP activity is expressed as mmol substrate hydrolyzed as related to unit time and kg body weight. Maltase activity was analyzed using maltose as substrate. Maltase activity is expressed as mmol substrate hydrolysed as related to unit time and kg body weight. For trypsin activity analyses, the frozen tissue with intestinal content was homogenized and suspended in dH2O. Trypsin activity was measured colorimetrically according to Kakade et al., using the substrate benzoyl-arginine-p-nitroanilide. Activity is expressed as change in optic density related to mg tissue homogenate. Amylase activity was measured using a Randox amylase assay kit 
by hydrolysis of benzylidene-blocked p-nitrophenyl maltoheptaoside. The amylase activity is expressed in mU g21 tissue homogenate. The bile salt concentration in samples was determined using the Enzabile kit from Nycomed Pharma AS Diagnostics and expressed as mg g21 tissue homogenate. salmon would be expected. Transcriptional expression of T helper cell marker CD4, the cytokines interleukin 17a and interferon gamma, proliferating cell nuclear antigen, and heat shock protein 70 in the DI tissue were assessed. Total RNA was extracted from dissected DI on the day 99 sampling from 16476508 three fish per tank, i.e. nine animals per diet. RNA purification and quality control, DNase treatment, complementary DNA synthesis, primer optimisation and quantitative PCR assays were performed as described in detail elsewhere. Quantitative PCR primers were obtained from the literature or designed using Primer3. See Calculations Histology Formalin-fixed liver, PI and DI samples from four fish per tank, i.e. 12 fish per diet, sampled on day 99 w