Role of globulins and albumins in oil-water interface and emulsion stabilization properties of pulse proteins

Pulse protein extracts have received considerable attention in food applications, for instance, as emulsifiers. These extracts are complex protein mixtures consisting primarily of globulins and albumins, but how these individual protein fractions affect the emulsifying properties is still poorly understood. In this study, we separated globulins and albumins from lentil, faba bean and chickpea, and systematically investigated their role in the stabilization of oil-water interfaces and emulsions. We identified key molecular and interfacial parameters of these pulse proteins (including the whole protein mixtures) which are important in emulsion stabilization. Pulse proteins with low denaturation enthalpy, a feature of the pulse albumins, tend to generate smaller oil droplets, most likely due to their small particle size (4–5 nm) which enables dense pack of proteins at the oil-water interface and promote the formation of stiff protein network structure that helps to stabilize the newly generated oil droplets during/after homogenization. High thermal stability, as observed in pulse legumin globulins, is detrimental to the interfacial stiffness, and the resistance of the interfacial network structure to large deformations due to the lack of protein-protein in-plane interactions and tend to give low emulsion stability under high shear blending. During emulsion storage, pulse albumins tend to cause droplet flocculation/coalescence also in the whole protein mixtures, likely due to their low surface charge and smaller particle size that could not provide enough electrostatic and steric repulson. Overall, albumins dominate the emulsion stability of pulse proteins under high shear blending, and globulins dominate the emulsion storage stability. This study may help to guide the screening of pulse proteins for desired emulsifying properties. The proposed key molecular and interfacial parameters could be used for further modeling, for example, based on artificial intelligence, to better understand and predict the functionalities of pulse proteins in multiphase systems.