Enzyme treated Lupinus spp. seeds as an alternative source of protein for broilers
The studies reported in this thesis were carried out to evaluate the effects of enzyme treatment on the nutritive value of three lupin species (L. luteus, L. albus and L. angustifolius) and evaluate their suitability as a source of protein for growing broilers. Some preliminary in vitro work on the effect of quinolizidine alkaloids on bacteria from poultry excreta was carried out. Furthermore the use of near infrared spectroscopy (NIRS) as an alternative technique for assessing the chemical composition and nutritive value (such as metabolizable energy and digestibilities) was investigated. Compositional analyses of the various lupins, fractions and diets were conducted. The chemical composition of the lupin seeds used in this study showed a high crude protein content and a variable content in ether extract and neutral detergent fibre. The alkaloid content of the seeds was determined and allowed the classification of the lupin into bitter (L. luteus cv Cardiga), semi-sweet (L. albus cv Estoril) and sweet (L. albus from France and L. angustifolius from Australia) seeds. Alkaloids from lupins decrease feed consumption and may affect the digestive capacity of the animal. A preliminary study on the effects of sparteine and alkaloids extracted from L. luteus seeds on bacteria from poultry excreta indicated that extracts of alkaloids from L. luteus had an inhibitory effect on coliform growth that was not, however, as great as that of sparteine. Sparteine, essentially prevented growth of coliforms at concentrations of 10 g dm-3. Lactobacilli were apparently not directly affected by the lupin extract of alkaloid or the isolated sparteine. This in vitro work suggests that it may be possible to influence the gut microflora in a beneficial manner by using the appropriate concentration of the appropriate alkaloids. A set of tube feeding experiments (by gavage) were carried out to evaluate the effect of different enzymes on the nutritive value of L. luteus (cv. Cardiga) and L. albus (cv. Estoril) seeds. The first experiment was a preliminary one which allowed the modification of the tube feeding assay in order to suit the characteristics of the test materials used in this study. The following six experiments tested the effects of pre-incubation, as a wet mash, of a polygalacturonase, five proteases, a pectinase and an [Special character omitted]-galactosidase at variable levels, on the nutritive value of L. luteus (cv Cardiga) and L. albus (cv. Estoril). The nutritive value of the lupin seeds was evaluated essentially by measuring metabolizable energy (ME) and amino acid (AA) digestibilities. Results indicated that L. albus seeds, irrespective of enzyme treatment, had a higher ME expressed as TMEn than L. luteus seeds. Carbohydrase containing enzyme preparations, caused variable improvements in the ME value of lupin seeds. Polygalacturonase caused a significant increase in the ME of L. luteus seeds, which may have been mediated by an increase in protein digestibility as shown by improvements in the AA digestibility. Increased concentrations of pectinase tended to be more effective in the improvement of ME of L. albus, while increasing concentrations of a-galactosidase were so for L. luteus. The fact that the carbohydrases acted differently upon the two lupin species was probably due to their different concentration in constituent carbohydrates. Pre-incubation with exogenous proteases decreased the ME of the lupin seeds. The reasons for the depression were not clear, however high concentrations of proteases, expressed as U kg-1 of lupin seeds, may have interfered with the gastro intestinal tract and also with the endogenous enzymes of the bird resulting in reduced utilisation of the pre-incubated material. A growth experiment was design to investigate the effects of the inclusion of L. luteus (cv. Cardiga), L. albus (from France) or L. angustifolius (from Australia) as the main or secondary source of protein in diets for growing broilers. Lupin seeds were either pre-incubated as a wet mash, or the lupin containing diets were supplemented with a carbohydrase, a protease or a mixture of both. Diets were fed to the animals over a three week period. The results obtained tended to support the hypothesis that the enzymes had little effect on the other dietary ingredients but had their main effects on the lupins in the diets.