The symbiosis with ruminal microorganisms has given ruminants the unique ability to utilize structural carbohydrates and non-protein nitrogen. On the other hand, the microorganisms can convert essential feed ingredients into low value nutrients, which has emerged the need for rumen protection technologies. Recently, a novel rumen protection technology using the enzyme polyphenol oxidase (PPO) has been developed within our research group for the protection of polyunsaturated fatty acids against ruminal biohydrogenation. Herein, linseed oil was emulsified in PPO-containing protein extract and cross-linking of the interfacial protein layer was induced by the addition of a diphenolic PPO substrate. The potential of this cross-linking technology for the protection of lipophilic compounds has been explored before. As there is also a need for protection of hydrophilic compounds, the first focus of the current thesis was to explore this technology towards the protection of both hydrophilic and lipophilic compounds by interfacial cross-linking of double emulsions (Chapter 2). A first requirement for protection of the hydrophilic compounds was the efficient encapsulation of such compounds in double emulsions stabilized by PPO extract, which was evaluated in several experiments. Unfortunately, PPO extract was found an unsuitable emulsifier to prevent the migration of hydrophilic compounds from the inner droplets of double emulsions. Therefore, the consequent part of the thesis focused on further optimization of the protection of polyunsaturated fatty acids by use of PPO. In the subsequent research chapter, it was shown that emulsions prepared by a variety of protein emulsifiers could be efficiently protected by addition of the PPO extract after emulsion preparation. This approach could be applied in the future to further exploit the potential of PPO-mediated cross-linking of double emulsions.

 In previous experiments, the PPO technology successfully allowed to protect linseed oil emulsions containing 2 wt% oil only. Accordingly, the aim was to optimize this technology to a more efficient use of the PPO extract by protecting higher oil levels with lower amounts of PPO extract. First, it was shown that 2 wt% oil emulsions could be protected by preparing the emulsions in absence of the PPO extract using other, more suitable, protein emulsifiers, and subsequent addition of PPO extract (80 wt%) and its substrate. Afterwards, a series of experiments was performed to evaluate the protection of linseed oil emulsions with higher oil levels and lower amounts of PPO extract, which resulted in the efficient protection of emulsions containing 8 wt% oil and only 60 wt% PPO extract i.e. 6.5 times more oil could be protected with the same amount of PPO extract as before.

To further elucidate essential factors for efficient ruminal protection the cross-linked protein fractions of emulsions with different levels of protection efficiency were studied (Chapter 4). The amount of protein-bound phenols in the adsorbed proteins was positively correlated with the protection efficiency against ruminal biohydrogenation, where an excess of phenols relative to the theoretically available binding sites of the protein seemed a prerequisite to obtain high ruminal protection efficiency. On the other hand, a higher amount of non-adsorbed protein in the emulsions was associated with a lower protection efficiency.

In a final Chapter (5A), data from several experiments were combined to build a model for the prediction of the ruminal protection efficiency from readily available emulsion characteristics. Several equations were fit to the training data but none of the models showed accurate prediction in external validation. Combination of results obtained in Chapter 4 and the quantitative model of Chapter 5A suggested the amount of total adsorbed protein-bound phenols might be the most appropriate predictor of protection efficiency but more data should be collected to confirm this hypothesis.

In conclusion, a new strategy for the rumen protection of feed compounds has been explored and further optimized. Additionally, some essential features of this technology have been identified, which can be a guideline in future exploration of this enzyme based protection technology.

Future objectives might be the formation of a dry product from the cross-linked emulsions, e.g. by spray-drying, because of practical limitations related to the supplementation of a liquid product and the potential problems with shelf-life. Additionally, post-ruminal release is an absolute requirement for development of rumen-bypass emulsions. Either the release of fatty acids from the emulsion can be studied directly or degradation of the crosslinked-protein matrix can be monitored, e.g. by SDS-PAGE. Finally, next to in vitro studies, the protection efficiency against rumen biohydrogenation, post-ruminal release and transfer to the milk should be evaluated in vivo to validate the effectiveness of the developed product and to examine the in vitro-in vivo relationship.