Plant-based cheese is one of the most challenging products within dairy alternatives due to the complexity of cheese’s physicochemical and sensory properties. Currently, plant-based cheese production mainly relies on solid fats and starch to obtain a firm texture and on flavouring agents to emulate cheese flavours, leading to products with low protein content that do not fulfil consumer sensory expectations. This PhD thesis aims to investigate the use of fermentation and enzymatic treatment to enhance texture and flavour in pea protein emulsion gels for plant-based cheese applications. The primary objective is to investigate the development of fermentation-induced pea protein emulsion gels with a firm texture, dairy-related volatile organic compounds, and reduced beany-related volatile organic compounds using lactic acid bacteria fermentation and transglutaminase.

To produce fermentation-induced pea protein gels, it is essential to comprehend the characteristics of the chosen raw materials before and during the fermentation process. The processing of these raw materials significantly impacts the physicochemical properties of proteins, influencing their capacity to form gels during fermentation. Additionally, the inclusion of oil in the matrix plays a role in structure formation, subsequently impacting the firmness of the resulting gel. In this thesis, pea protein isolate emulsions with 10% protein and varying liquid oil levels remained stable over fermentation time, indicating their suitability for fermentation-induced gels. Increasing oil content resulted in weaker gels with reduced elastic modulus after seven days of storage under refrigeration in closed containers. The findings suggest that an optimal pea protein matrix for fermented gels can be produced with 10% protein and 10% oil, which was established as the starting model system for the different work packages in this PhD thesis.

Novel plant lipid sources such as oleosomes and oleogels are potential alternatives to coconut fat in plant-based cheese. They were integrated into the previously designed model system, and fermentation-induced pea protein gels were produced. These were then compared to a traditional dairy fresh cheese and to model systems, including liquid oil or coconut fat, to assess the effects of oleosomes and oleogels on emulsion stability and gel structure. The study revealed that the choice of lipid source did not significantly affect emulsion stability. While the dairy cheese exhibited higher hardness than its pea protein-based counterparts, as determined by texture analysis, there were no significant differences in the rheological properties of pea protein gels containing oil, oleosomes, or oleogels compared to the reference dairy cheese. This suggests that, while the protein network plays a primary role in determining gel firmness, it is feasible to produce fermentation-induced pea protein gels with liquid oil, oleosomes and oleogels that could closely mimic the rheological properties of traditional dairy fresh cheese.

Fermentation is a powerful tool to enhance plant-based cheese’s textural properties and flavour profiles. However, it is crucial to tailor bacterial cultures to match the desired texture and flavour attributes specific to cheese production, which may diverge from those needed in other plant-based products such as yoghurt. In pursuit of this goal, two parallel screenings were conducted to assess the competencies of lactic acid bacteria (LAB) for fermentation-induced pea protein gels. In the first one, twenty-four different bacterial blends involving four commercially available starter cultures and six adjunct combinations were designed. Their acidification performance, impact on gel firmness, and capacity to mitigate undesirable off-aromas while generating dairy-related aromas were assessed through pH measurement, compression tests, and volatile organic compound (VOC) analysis, respectively. Notably, blends featuring Vega™Harmony (Streptococcus thermophilus, Lactobacillus delbrueckii subsp. bulgaricus, Lactobacillus acidophilus, Bifidobacterium lactis, Lacticaseibacillus paracasei) and Vega™Classic (S. thermophilus, Lb. bulgaricus) exhibited significantly faster acidification and enhanced efficacy in mitigating compounds responsible for green and beany off-aromas commonly found in pea-based matrices. Furthermore, the analysis revealed the presence of dairy-like compounds in all fermented samples, with variations in their production levels contingent upon the specific bacterial blend inoculated. In the second screening, a pool of 90 LAB strains was evaluated, and 13 were selected based on their performance in acidification, gel firmness enhancement, beany-related VOCs reduction, and dairy-related VOCs production. These selected strains were combined into 64 different blends of five each and re-evaluated for the mentioned parameters. Blends containing S. thermophilus, Lactobacillus, and Lactococcus strains significantly reduced beany-related aroma compounds, improved the dairy-related aroma profile, and enhanced gel firmness. These two screenings allowed for the identification of strain combinations that can pave the way to designing starter cultures tailored for plant-based cheese.

Pea protein emulsion gels produced through fermentation present firm structures, but their hardness and springiness can be improved with enzymatic treatment with transglutaminase (TG). The impact of this enzyme on the microstructure, texture, and rheological properties of fermentation-induced pea protein emulsion gels was evaluated while also examining the influence of storage time. The presence of TG in the gels reduced gel porosity, increased storage modulus, and improved textural properties, yielding firmer and more springy gels. Although both TGtreated and untreated gels showed increased porosity during storage, the effects of time on texture and rheological behaviour were limited, indicating minimal structural changes once the gels formed. The absence of TG led to a more remarkable coalescence of oil droplets within the gel matrix after 16 weeks. These results highlighted the benefits of TG treatment for enhancing the textural and rheological properties of fermentation-induced pea protein emulsion gels.

n addition to exploring textural improvements in fermentation-induced pea protein emulsion gels for plant-based cheese, aroma and peptide profile development over time were also investigated. The impact of the top-performing four bacterial blends from the previously described screenings on volatile and non-volatile metabolites was evaluated over different storage periods of up to 16 weeks. Moreover, the presence of TG and its effect on the VOCs and peptidomic profiles of fermented pea protein emulsion gels were examined. The different bacterial blends greatly impacted the production of VOCs, and the effect of time differed depending on the inoculated bacterial blend. Additionally, the presence of TG did not affect the VOCs profiles, regardless of the inoculated blend. However, TG did affect the non-volatile profile of the samples. In samples fermented by three of the four inoculated blends, more bitter-associated peptides were detected when TG was absent, although the difference was not large. In contrast, in samples fermented by the fourth blend (which only contained S. thermophilus and Lb. bulgaricus), a higher number of bitter peptides was detected in the presence of TG. These results underscore the need for further studies that ensure that by improving texture with TG, the flavour profile is not worsened by the presence of this enzyme.

This thesis provides valuable insights into critical considerations for the production of fermentation-induced pea protein emulsion gels for plant-based cheese. These considerations include the design of a starting matrix, the inoculated optimal bacterial blends, and strategies to enhance and improve texture and aroma profiles. The findings suggest using a pea protein model system, developing bacterial blends specifically tailored for plant-based cheese applications, and employing the synergistic application of TG with LAB blends for improved textural properties. Furthermore, preliminary data on the development of VOCs over time is provided and continuous research is encouraged to help understand how to improve aroma and flavour during plant-based cheese ripening. These research outcomes contribute to understanding fermentation in plant-based applications and pave the way for future investigations in this field, ultimately facilitating the production of high-quality plant-based cheeses.