Protein Yield and Protein Isolate Quality when applying pH-shift Processing on Cod, Haddock and Blue Whiting Fillets- Effects of Replacing Centrifugation with Filtration

A more sustainable fishery and a better utilization of the harvested fish raw material are currently of great importance to prevent overexploitation of our marine resources. Production of protein isolates to use for example as surimi has for many years been an important way of better utilizing low-value fish species and fish rest raw materials. In the pH-shift protein isolation technique, fish muscle proteins are solubilized in water at strong acid or alkaline conditions, which allows for removal of non-soluble materials like bones, skin, cartilage and fat in a first separation step. The purified solubilized proteins are then precipitated at pH 5.5 and are recovered as a protein isolate in a second separation step. Traditionally, the two separation steps are carried out with centrifugation, a method that can be very costly in large scale. Also, several previous studies have reported on the formation of large sediments in the first centrifugation step containing not only non-soluble materials, but also nearly solubilized and solubilized proteins. These sediments are hard to recover, and reduces the final protein yields. The main aim of the present thesis was therefore to evaluate filtration as an alternative to centrifugation in the first and a second separation step of the acid and alkaline versions of the pH-shift process. The responses were total protein yield, basic composition of the protein isolates and quality of surimi gels produced from the isolates. The raw materials tested were fresh cod and haddock as well as fresh and frozen blue whiting. Using filtration instead of centrifugation in the first separation step significantly increased the protein yields; for cod and haddock with 26% and for blue whiting with 38-60%. Changing separation method in the second step did not significantly influence the protein yield. The main compositional difference between isolates produced with filtration instead of centrifugation in the first separation step was a higher lipid content with the former, most likely as filtration separates according to particle size instead of density. Further, surimi gels derived from the filtration based process were in most cases weaker and less elastic than gels derived from the centrifugation based process. Slightly lower protein content in the surimi gels caused by a higher lipid content could be one reason. Another could be the slightly higher levels of very high molecular weight myofibrillar proteins like titin in isolates produced with centrifugation. For both isolates and surimi gels there was a tendency towards higher whiteness values when acid solubilization and centrifugation-based separation in the first step was used. In conclusion, the filtration based pH-shift method is very promising alternative when protein yield is in focus. The impact of the higher lipid content and lower gelation capacity will be linked to the final use of the protein isolates. Many applications do not include a requirement for gelation