Pea grains may partially replace soybean or rapeseed meals and cereals in ruminant diets, but substitution by unprocessed peas is limited by high ruminal protein solubility. The effect of combined ensiling and toasting of peas using a mobile toaster (100 kg/h throughput rate, 180 to 190 °C supplied air temperature) on rumen-undegraded protein (RUP) was tested in vitro using the Streptomyces griseus protease test. The effects of ensiling plus toasting on apparent digestibility of organic matter (OM), gross energy (GE), and proximate nutrients were examined in a digestion trial. Concentrations of metabolizable energy (ME) and net energy lactation (NEL) were calculated. Native peas had 38 g RUP/kg dry matter (DM), which was 20% of crude protein (CP). Rumen-undegraded protein increased three-fold after ensiling plus toasting (p < 0.001). Acid detergent insoluble protein increased five-fold. Apparent digestibility was 0.94 (OM), 0.90 (CP), and above 0.99 (nitrogen-free extract, starch, and sugars) and was not altered by the treatment. The ME (13.9 MJ/kg DM) or the NEL (8.9 MJ/kg DM) concentration was similar in native and ensiled plus toasted peas. This technique can easily be applied on farms and may increase RUP. However, it needs to be clarified under which conditions pea protein will be damaged.

No abstract

No abstract

Effective degradation (ED) of crude protein (CP) was estimated in vitro at 0.02, 0.05 and 0.08 h−1 assumed ruminal passage rates for a total of 40 feedstuffs, for which in situ ED was available and used as reference degradation values. For this, the Streptomyces griseus protease test was used. The differences between in vitro CP degradation and the in situ CP degradation values were lowest in legume grains and highest in cereal by-products and barley. The differences between in situ and in vitro ED were expressed using a degradation quotient (degQ), where degQ = (EDin vitro − EDin situ)/EDin situ. Among the tested feedstuffs, eight specific clusters were identified according to degQ for the assumed passage rates. The feedstuffs clustered in an unspecific way, i.e., feedstuffs of different nutrient composition, origin or treatment did not necessarily group together. Formaldehyde–treated rapeseed meal, soybean meal, wheat, a treated lupin, sunflower meal and barley could not be assigned to any of the clusters. Groupwise degradation (range of degQ for assumed passage rates are given in brackets) was detected in grass silages (−0.17, −0.11), cereal by-products together with sugar beet pulp (−0.47, −0.35) and partly in legume grains (−0.14, 0.14). The clustering probably based on different specific nutrient composition and matrix effects that influence the solubility of feed protein and limit the performance of the protease. The matrix can be affected by treatment (chemically, thermally or mechanically), changing the chemical and physical structure of the protein within the plant. The S. griseus protease test had reliable sensitivity to reflect differences between native feedstuffs and treatments (thermally or chemically) that were found in situ. The in situ results, however, are mostly underestimated. The clustering results do not allow a clear conclusion on the groupwise or feed-specific use of carbohydrate-degrading enzymes as pre- or co-inoculants as part of the S. griseus protease test and need to be tested for its potential to make this test more conform with in situ data.