Of proteins and processing: mechanisms of protein damage upon rapeseed processing and their effects on nutritional value

Hydrothermal processing is a common practice during the manufacture of protein-rich feed ingredients, such as rapeseed meal (RSM), and feeds. This processing step can induce physical and chemical changes to the proteins, thereby reducing the digestibility and utilization of crude protein (CP) and amino acids (AA). Whilst most literature has linked the chemical changes to the proteins with negative effects on protein digestibility, the effects of the physical changes of the protein conformation have not been considered simultaneously. Hence, the aim of this thesis was to provide further insight into the mechanisms of protein damage during ingredient/feed processing and their effects on protein hydrolysis/digestibility. In Chapter 2, the available literature on the physical changes that occur to vegetable proteins used in swine diets after processing was reviewed. Overall, hydrothermal processing increases the contents of intermolecular/intramolecular β-sheets and disulfide bonds, which were negatively correlated to protein digestibility. The correlations, however, were dependent on the type of protein analysed. When the physico-chemical changes in the proteins occur during processing of the ingredients, proteins usually become less responsive to further processing treatments. Rapeseed proteins were used as model, as this oilseed is hydrothermally processed during the oil extraction process and is further processed when incorporated in animal diets.

Protein damage during production of rapeseed meal

There is high variability in the nutritional value of commercial RSM. The variation is mainly due to the conditions used during the desolventization/toasting step. Therefore, the aim of the experiment in Chapter 3 was to characterize the secondary structure and chemical changes that occur during toasting of RSM and their effects on in vitro protein digestibility. A cold defatted RSM was toasted for 120 min with samples obtained every 20 min. Increasing the toasting time from 0 to 120 min increased protein denaturation by 3-fold and decreased protein solubility by 4-fold, lysine content by 23% and the reactive lysine content by 37%. The proportion of intermolecular β-sheets increased after the initial 20 min of toasting, but steadily decreased thereafter. The contrary was observed for the proportion of α-helices. The changes in the secondary structure of proteins were not correlated to the rest of the physical and chemical changes. Therefore, changes in the secondary structure of proteins cannot be considered good indicators of damage to proteins due to hydrothermal processing. The rate of protein hydrolysis decreased by 2-fold when toasting time was increased from 0 min to 120 min. The changes in protein solubility and lysine/reactive lysine contents were positively correlated to the rate of protein hydrolysis. Changes to the physical conformation of rapeseed proteins occur at faster rates during toasting compared to chemical changes.

In Chapter 4, it was hypothesized that the decrease in the rate of protein hydrolysis with increasing toasting time was due to the reduction in protein solubility. In order to test this hypothesis, the soluble and insoluble protein fractions from each of the RSM studied in Chapter 3 were separated and hydrolysed. Hydrolysis kinetics and the molecular size distribution of the peptides resulting after hydrolysis were analysed. The extent of hydrolysis of the insoluble protein fraction was 44% higher than that of the soluble protein fraction. The rate of hydrolysis of the soluble protein fraction separated from the hydrothermally treated RSM was 3-9 fold higher than that of the insoluble protein fraction. In the insoluble fraction, formation of both disulfide bonds and Maillard reaction products (MRP) (fructosyl-lysine [FL], carboxymethyl-lysine [CML] and carboxyethyl-lysine [CEL]) was noticed, which explains the decrease in the rate of protein hydrolysis with longer toasting times. Overall, increasing the toasting time of the whole RSM and the insoluble protein fraction increased the size of the peptides resulting after enzymatic hydrolysis. A shift in the mechanism of protein hydrolysis from a more one-by-one type to a more zipper-type likely explained the correlations between the rate of hydrolysis and the molecular size distribution after hydrolysis. Protein solubility seems to be a key parameter for understanding the decrease in the rate of protein hydrolysis with increasing toasting time.

The correlations between two in vitro protein digestibility methods and the standardised ileal digestibility in growing pigs of severe thermally-treated soybean and rapeseed meals were studied in Chapter 5. Soybean meal and RSM were toasted in the presence of lignosulfonate in order to induce severe thermal damage to the proteins. In vitro protein digestibility was analysed using the two-step enzymatic method (pepsin at pH 2.0 and pancreatin at pH 6.8) and the pH-STAT method. The standardised ileal digestibility values were obtained from a previous experiment, in which ileal-cannulated growing pigs were used. The degree of hydrolysis after 10 min was positively correlated (r = 0.95, P = 0.046) to the standardised ileal CP digestibility. The in vitro rate of protein hydrolysis using the pH-STAT method and CP digestibility using the two-step enzymatic method tended to be positively correlated to the standardised ileal digestibility of CP (r = 0.91, P = 0.09, for both in vitro methods). In conclusion, both in vitro methods might be used for the in vivo digestibility of severe thermally-treated ingredients.

Effects of diet processing on protein digestibility of RSM with different extents of damage

The processed ingredients (e.g. rapeseed meal) are mixed with other ingredients and processed further during the compound feed manufacturing process. The effects of the diet processing methods (e.g. pelleting and extrusion) on protein digestibility could depend on the extent of the damage of the ingredients used. The aim of Chapter 6 was to test the effects of toasting time of rapeseed meal, diet processing method and the interaction between both on protein digestion along the gastrointestinal tract and apparent/standardised ileal digestibility of CP and AA. Mash, pelleted and extruded diets were manufactured using either 0, 60 or 120 min toasted RSM as the only protein source, for a total of 9 different experimental diets. Whilst increasing the toasting time decreased the contents of lysine and reactive lysine in the diets, no effects were noticed after pelleting or extrusion of the diets compared to the mash. The mean particle size of the diets was reduced from 479 μm in the mash diets to 309 and 211 μm after pelleting and extrusion, respectively. A total of 81 growing boars were individually fed with one of the experimental diets. Following euthanasia, the small intestine was divided in 3 sections of equal length and the contents of the final 1.5 m of each small intestine section were sampled. The apparent CP digestibility for each section of the gastrointestinal tract was used to calculate the rate of CP digestion based on a second order equation. The rate of digestion was higher in the diets containing 0 min toasted RSM compared to the diets that contained 60 or 120 min toasted RSM. The diet processing method tended to affect the rate of protein digestion, with higher rates for the extruded > pelleted > mash diets. Significant effects of the interaction between toasting time and diet processing method were found on the apparent ileal CP digestibility. Whilst a lower apparent ileal CP digestibility was found in the 120 min toasted RSM mash diet compared to the 0 and 60 min toasted RSM mash diets, no differences were observed between the different toasting times in the pelleted and extruded diets. Similar significant interactions were noticed for the apparent ileal digestibility of some dispensable and indispensable AA (e.g. arginine, isoleucine, leucine, methionine, threonine, alanine, glycine, proline, serine). Pelleting of the 60 and 120 min toasted RSM diets increased the standardized ileal digestible CP content by 6 and 15%, respectively, compared to the 60 and 120 min toasted mash diets. Extrusion of the 0, 60 and 120 min toasted RSM diets increased the standardized ileal digestible CP content by 5, 9 and 12%, respectively, compared to the 0, 60 and 120 min toasted mash diets. Similar positive effects of pelleting and extrusion were obtained for the apparent ileal digestible contents of lysine and reactive lysine, especially for the diets that contained RSM toasted for longer times. In conclusion, the severe effects of protein damage during the production of RSM on protein digestibility can be (partially) ameliorated by processing of the diets.

Processing of ingredients and diets can lead to the formation of early (e.g. FL) and advanced (e.g. CML and CEL) MRP. These MRP have been associated with common metabolic disorders, for example atherosclerosis. Absorption of dietary MRP has been previously estimated based on indirect measurements, such as concentrations in blood, urine and faeces, which could be biased by endogenous formation of MRP and deposition in tissues. Hence, the aim of Chapter 7 was to measure the apparent ileal digestibility of early (FL, determined as furosine after acid hydrolysis) and advanced (CML and CEL) MRP. The same diets and ileal digesta samples as in Chapter 6 were used in this study. The 0 min toasted RSM diets (mash, pelleted and extruded) were excluded from this study. Sucrose was added to these diets immediately before feeding in order to increase their feed intake, which could confound the determination of the apparent ileal digestibility of the MRP for these treatments. The content of FL was higher in the 120 min compared to the 60 min toasted RSM diets, whereas it was lower in the extruded diets compared to the pelleted and mash diets. The decrease in FL content after extrusion of the diets can probably be related to conversion of the early into advanced MRP, as the content of CML was higher in the extruded diets compared to the pelleted and mash diets. The contents of FL, CML and CEL were positively correlated to the contents of these compounds in the ileal chyme. The apparent ileal digestibility of FL, CML and CEL for the different diets ranged from -8.5 to 19.1%, -0.2 to 18.3% and 3.6 to 30%, respectively. In conclusion, the apparent ileal digestibility of the early and advanced MRP from thermally-treated RSM diets in growing pigs were overall low and did not seem to be related to the contents of these compounds in the diets.

The results of this thesis indicate that the changes to the physical conformation of proteins during toasting of RSM occur at a faster rate than chemical changes. Both types of changes affect protein solubility, which is important in determining the rate of protein hydrolysis. It is suggested in this thesis that the rate of protein hydrolysis of hydrothermally-processed ingredients is probably the main contributing factor for the in vivo protein digestibility. This was demonstrated, as increasing the toasting time of RSM decreased the rate of CP digestion in growing pigs. The negative effects of longer toasting times on CP and AA digestibility, however, could be ameliorated by pelleting and extrusion of the diets. The positive effects of diet processing methods (pelleting and extrusion) on the digestibility of damaged proteins from thermally-treated ingredients should be taken into account in feed evaluation studies and formulation of practical diets.