Optimization of protein recovery from Atlantic salmon (Salmo salar) head and backbone by response surface methodology and characterization of functional properties and nutritional value

Atlantic salmon (Salmo salar) head and backbone by-products were hydrolysed using non-commercial protease enzyme (ERM 1) to produce protein. Response surface methodology was used to optimise conditions, including hydrolysis time, hydromodule and enzyme–substrate (E:S) ratio for maximum protein recovery. Highest protein recovery was obtained after 4 h hydrolysis, 1 L/kg hydromodule, and 0.39% of E:S ratio for the salmon head. Similarly, 3.75 h of hydrolysis time, 2.67 L/kg of hydromodule and 0.499% of E:S were found optimal for the salmon backbone. Total amino acid (TAA) composition revealed the presence of all essential amino acids in both hydrolysates. The sum of 16 TAAs was approximately both in salmon head and backbone samples 70 g/100 g, while FAAs were much higher in salmon head (13.4 g/100 g) then in the salmon backbone (8.8 g/100 g). The hydrolysates prevented the growth of E. coli K-12, but no significant effect on Listeria innocua (ATTC 33090) growth was seen. Fish hydrolysates showed nitrogen solubility indices above 90% at pH 5–8, with one exception of the salmon head hydrolysate at pH 5 with a value of 67.8%. Samples formed gels at 5 and 10% protein concentration. Gels were weak compared to gelatine gels.