The influence of gestation feeding strategy on body composition of gilts at farrowing and response to dietary protein in a modified lactation

Previous experiments have indicated that reproductive function in lean, mod ern genotypes may be more dependent on body protein mass than, as previousl y believed, on body lipid reserves. This was investigated in a 3 x 2 factor ial arrangement of treatments, involving 60 first-parity sows, comparing th ree pregnancy feeding strategies and two lactation diets. During pregnancy, sows were fed either a basal diet (5 g lysine/kg, 13 MJ of DE/kg [CI) or t he same quantity of basal diet + energy source [E], or additional basal die t supplying both protein and energy [A]. The level of supplement for E and A was adjusted weekly to achieve a backfat thickness measurement (P-2 posit ion) of 28 mm at farrowing. Isoenergetic lactation diets were fed to appeti te and provided either high (180 g CP/kg, 9 g lysine/kg [HI) or low lysine (120 g CP/kg, 6 g lysine/kg [L]). From d 21 of lactation, sows were separat ed from their litters and housed next to a boar for 8 h each day; final wea ning occurred on d 31. Pregnancy treatment differences in backfat and weigh t were achieved, with C sows having less backfat on d 1 of lactation than E and A sows (E = 28.1, A = 28.0, C = 22.7 kg, P < 0.001). Sows fed addition al basal diet were heavier than E sows, which were heavier than C sows (E = 190, A = 201, C = 178 kg, P < 0.001). Average feed intake over lactation s howed a pregnancy feeding effect, with E sows eating less than A or C sows (E = 4.9, A = 5.2, C = 5.4 kg/d, P < 0.005). Total lactation weight loss wa s affected by pregnancy feeding (E = 18.0, A = 19.0, C = 8.4 kg, P < 0.05) and by lactation diet (L = 19.0, H = 11.3 kg, P < 0.05), whereas total lact ation backfat loss was affected only by pregnancy treatment (E = 6.9, A = 6 .5, C = 4.6 mm, P < 0.05). No pregnancy treatment or lactation diet effects were observed for litter performance. Lactation diet affected weaning-to-e strus interval, with more sows on the H diet coming into estrus within 6 d of partial weaning (P < 0.05), but there was no pregnancy treatment effect. Therefore, voluntary feed intake during lactation was suppressed by increa sed fat reserves at a limited body protein mass but not when body protein m ass was also increased. Partial weaning-to-estrus interval was increased by reduced dietary protein.