Pea protein functionality: Tailor-made through fractionation

For sustainability reasons, there is an ongoing shift from animal- to plant-based proteins, which is often referred to as the protein transition. The research of this PhD thesis was conducted in the context of that protein transition. The PhD thesis describes the effect of mild aqueous fractionation processes on the functionality of pea protein, in terms of molecular and microstructural characteristics. The thesis is structured according to the type of model foods studied.

The first part focusses on pea protein dispersions and gels. The first step in a protein fractionation process is milling of the seed. The morphology of the pea seed before and after milling was visualized using electron microscopy. Subsequently, alkaline extraction and isoelectric precipitation was applied to obtain five pea protein fractions exposed to different extents of processing and with different compositions. It was found that extensively fractionated pea proteins had a substantial thickening capacity in dispersion. This physical behaviour was attributed to the ability of pea protein to form aggregates with a high specific volume. Furthermore, pea proteins were able to form homogeneous coacervate dispersions at certain pH and salt conditions, given that the proteins were only mildly fractionated. Upon heating, mildly fractionated pea protein fractions were able to form stiff and ductile gels, while more extensive fractionation resulted in pea fractions that formed soft and brittle gels. The first part of this thesis shows that with mild fractionation specific pea protein functionalities are preserved.

The second part of this thesis reports on the stabilization of oil and air after incorporation in foams, emulsions and emulsion-filled gels. Fractionation could be used to induce a separation between pea albumins and globulins, and it was found that albumins showed good foaming properties, while globulins were better emulsifiers. The foaming properties were related to the interfacial rheology of albumins, and it was found that albumins were able to form a stiff layer at the air-water interface. Also, the ability of pea proteins to form emulsion-filled gels was investigated. It was found that different gel properties could be obtained when isoelectric precipitation was replaced by membrane filtration upon fractionation. At neutral pH, isoelectric precipitated emulsion-filled gels were heterogeneous and softer, while membrane filtrated emulsion-filled gels were homogeneous and stiffer. At pH 5 however, the emulsion-filled gels from the two pea protein fractions showed quite similar rheological behaviour. Another conclusion from the work on emulsion-filled gels was that oil did not reinforce any of the pea protein emulsion-filled gel structures. It was hypothesized that the lack of oil reinforcement could be attributed to weak interactions between the protein matrix and proteins at the interface.

The third part addresses hybrid plant-dairy protein dispersions and gels. Also here, it was found that fractionation processes could be optimized in such a way that pea proteins behaved like dairy proteins (i.e., whey protein) in terms of viscosity, solubility and gel stiffness. Membrane filtrated pea protein resembled the properties of whey protein most closely, which was primarily attributed to the unaggregated state of these pea proteins. However, upon partial replacement of whey protein by pea protein, it was found that the fractionation method did not matter, as mixtures of whey protein with isoelectric precipitated, membrane filtrated, and commercial pea protein isolate all behaved similar. When substituting whey protein with mildly processed fractions, significant differences were seen for different pea protein fractions. Pea globulin fractions could substitute part of the whey protein without affecting gel stiffness, while the pea albumin fraction was even able to enhance whey protein gel stiffness. This enhanced stiffness in the pea albumin - whey protein mixture was attributed to additional disulphide bonding upon gelation. Moreover, the gelation behaviour of the pea albumins and whey protein mixture was least affected by salt and pH differences.

Overall, it was found that a broad range of functional properties could be obtained with only one raw material, showing the versatility of pea as a protein source. Functionality changes due to fractionation could be related to changes in the composition and state of pea proteins in the various fractions. This also led to the realization that fractionation can be tailored to specific pea protein functionality. The thesis ends with highlighting the fact that potential application of tailor-made pea protein fractions, also require more insights into other factors, such as sustainability, nutritional value, and organoleptic properties of pea ingredients. The knowledge generated in this research and described in this thesis may facilitate further research on these factors and may contribute to the development of highly functional pea protein ingredients.