This study examines the impact of Lupinus angustifolius variety (C) and inclusion level (L) in broiler diets on the nutritional value, viscosity of ileal digesta and activity of gut microbiota. The experiment was conducted on 154 female 21-day-old broilers, allocated to 11 groups (kept individually). A reference lupin-free diet and 10 test diets containing one of five lupin seeds; Kadryl, Regent, Dalbor, Bojar and Tango, mixed with the reference diet at a ratio of 25 : 75 or 32 : 68 dry matter (DM) (low or high level of inclusion) were prepared. Diets were fed for 6 days, excreta were collected over last 4 days. Apparent metabolizable energy corrected to zero N balance (AMEN) of diets and AMEN of lupin seeds were calculated. Birds were sacrificed, ileal and caecal digesta were pooled by segments from two birds, and the activity of bacterial enzymes was determined. The ileal digesta viscosity was measured immediately (ileal viscosity immediate (IVI)) or after 6 days storage at −18°C (ileal viscosity frozen). AMEN of test diets were lower than the reference diet. Lupin AMEN values ranged from 6.04 MJ/kg DM for Regent at high level to 9.25 MJ/kg DM for Bojar at low level. High inclusion level numerically decreased AMEN value in all cultivars, except for Kadryl, for which it increased (significant C×L interaction). The IVI value was 2.6 mPa·s in the reference group, but ranged from 6.3 to 21.7 mPa·s in lupin-fed birds. It increased significantly with level for Regent, Dalbor and Tango but not for the other two cultivars (significant C×L interaction). There was a negative correlation between IVI and: apparent total tract N retention, fat digestibility from test diets, AMEN of diets and lupins. Ileal viscosity immediate was positively correlated with the activity of ileal α- and β-glucosidase and negatively with ileal α-galactosidase and caecal α-glucosidase. Ileal viscosity frozen ranged from 3.2 to 5 mPa·s and it was not correlated with lupins AMEN. This suggests that the digesta viscosity caused by narrow-leafed lupin is detrimental to its nutritional value and interfere with the gut microbial activity. In addition, the lupins viscosity was measured by two in vitro methods: the water extract viscosity (WEV) method and after incubation in conditions imitating in vivo digestion (enzyme-treated extract viscosity (EEV)). In vivo viscosity was weakly reflected by in vitro measurements as there was no correlation between IVI and WEV or EEV. Overall, findings suggest that the different cultivars of narrow-leafed lupin may have different value for practical application in broiler diets.

The effects of raw sweet lupin (Lupinus angustifolius) meal and its fractions on the growth and N utilization of rats were determined in two NPU and five N balance experiments. Sweet lupinseed grown in Western Australia, obtained as meal, unsupplemented (LMU), or fully supplemented with required amino acids (360 g/kg diet) (LMFS) was tested. In addition, six fractions were tested: aqueous non-dialysed extract at pH 7·0 (LPAND), dialysed extracts soluble (LPAD) and insoluble at pH 7·0 (LPADI), buffer-soluble extract at pH 7·0 (BUSOL), buffer-insoluble extract after dialysis at pH 7·0 (BUDI) and the residue (LMR) containing most of the material from meal insoluble in water and phosphate-citrate buffer. All diets based on fractions contained the same amounts of energy and protein and were supplemented with amino acids, vitamins and minerals to target requirements. Body N and lipid contents of rats fed on LMU and LMFS were reduced significantly in comparison with rats fed on positive lactalbumin (LACT) and non-protein diets (NPC) as negative controls. This wasdue in part to the lower retention of the absorbed N. As a result, the NPU and the biological value (BV) of sweet lupinseed proteins were less than expected. Urea-N outputs of the LMU- and LMFS-fed rats were also elevated. In contrast, true N and DM digestibilities of rats fed on LMU and LMFS were not significantly affected by the difference in the energy content of the diet. The replacement of lactalbumin in thediet with LPAND (196 g/kg), LPAD (148 g/kg), LPADI (124 g/kg), BUSOL (136 g/kg) or BUDI (119 g/kg) reduced dry body weight, N and lipid contents, NPU and BV compared with those obtained from the LACT control, even though the N and DM digestibilitieswere not significantly different. Inclusion of the residue fraction (170 g LMR/kg) had no apparent effect on any of the variables studied. Since sweet lupinseed had asmall amount of non-reactive lectin and LMR had some undesirable side-effects in these rats, it appears that the low nutritional value of LMFS for rats (NPU 0·62) despite the very high level of digestibility of its N, results from disturbances in N metabolism, and particularly from the low retention value of the absorbed N