Soybean meal (SBM) is an important protein source in swine diets, but inclusion is sometimes reduced and SBM is partially replaced with synthetic amino acids (AA) because this may sometimes reduce diet costs. Over the past few decades, reproductive efficiency in sows has improved significantly, due to genetic selection, resulting in litters that can exceed 20 piglets. This increase in prolificacy has, therefore, increased the metabolic demands for AA and energy during both gestation and lactation. Adequate energy intake in gestation supports fetal growth and mammary development, but excessive energy intake during gestation may reduce feed intake during lactation, which has a negative effect on milk production. Likewise, high protein intake during gestation will improve milk production, protein accretion, and litter and pig weight at weaning, but it may also lead to an increase in fat accretion rather than protein accretion in gestating sows.

Although synthetic AA may support growth performance, protein synthesis can only occur when all required AA are simultaneously available in the cell. Synthetic AA may be absorbed more rapidly than AA from intact proteins, which may result in early oxidation of synthetic AA before AA from intact protein arrives at the cell, limiting protein synthesis. However, it has not been demonstrated if low-protein diets containing synthetic AA can support the same reproductive performance, litter growth, and immune status of sows as feeding sows diets that contain more SBM and no synthetic AA. An experiment was, therefore, conducted to test the hypothesis that feeding sows diets based primarily on corn, SBM, and no crystalline AA will result in improved reproductive performance and immunity of sows compared with sows fed diets with less SBM and more corn and crystalline AA.

Experimental design

Two gestation diets and two lactation diets were formulated to meet estimated requirements for AA and other nutrients by gestating and lactating sows (NRC, 2012). Within each phase of production, one diet was a high-protein diet in which AA were furnished by corn and SBM, and the other diet was a low-protein diet in which the inclusion of SBM was reduced and synthetic AA were included to meet requirements for gestating or lactating sows. Both gestation diets also contained soybean hulls. The high-protein gestation and lactation diets contained 17.00% and 24.34% SBM, respectively, whereas the low-protein gestation diet contained 7.65% SBM, and the low-protein lactation diet contained 12.60% SBM.

Animals, housing, feeding, and sample collection

A total of 193 Camborough gilts and sows with initial body weight 190.04 ± 26.8 kg were used. Sows and gilts were allotted to eight blocks of 20 to 28 gilts and sows, using a randomized complete block design. Animals were allotted to experimental diets on the day of breeding, and feeding of gestation diets started on the day of breeding and continued until day 104 of gestation. At this time, sows were moved to the lactation facility, and feeding of the lactation diets was initiated. During the gestation period, daily feed allotments were provided at 0600 h. Feed allowance was 1.5 times the maintenance requirement for metabolizable energy (ME) for gestating sows (i.e., 100 kcal ME/kg body weight^0.60; NRC, 2012). From the 193 animals, 90 sows (parity 2 to 6) were placed in individual metabolism crates from day 45 to 56 (i.e., mid-gestation) with 12 sows in blocks one through six (i.e., six sows per treatment in each block) and nine sows in block seven and eight (i.e., four or five sows per treatment in each block), for a total of 44 and 46 sows for the high-protein and low-protein diets respectively. Crates were equipped with a self-feeder, a nipple waterer, and a fully slatted floor to allow for total, but separate collection of urine and fecal materials. Sows stayed in the crates for 11 days. The initial three days were considered the adaptation period to the crates, whereas urine and fecal materials were collected from feed provided during the following five days according to standard procedures using the marker-to-marker approach. On day 104, 154 sows were moved to the lactation unit and housed individually in farrowing crates. Sows were fed as in gestation from entry to the farrowing unit until farrowing, but diets were provided on an ad libitum basis from farrowing until weaning day. Sow body weights were determined, and also the number and body weight of pigs born alive, the number of mummies, stillborn pigs, and total pigs per litter after cross-fostering. Pigs were weighed again at weaning. Blood, colostrum, and milk samples were collected from the lactating sows, and pigs were weaned on d 20 after farrowing.

Results

Results indicated that daily nitrogen intake, nitrogen excretion in feces and urine, absorbed nitrogen, and retained nitrogen (g/d) were greater (P < 0.05) in gestating sows fed the high-protein diet compared with sows fed the low-protein diet (Table 1). However, nitrogen retention (% intake) was not different between treatments. Rectal temperature 24 h after farrowing of sows fed the low-protein diet was greater (P < 0.05) compared with the high-protein diets (Table 2). The number of live-born and total born pigs was not different between treatments, but sows fed the high-protein diet tended to produce less (P < 0.10) mummified pigs than sows fed the low-protein diet. Litter weights at birth and weaning were not different between treatments. Malondialdehyde was greater (P < 0.05) in sows fed the low-protein diet, but serum glutathione peroxidase and white blood cell count were greater (P < 0.05) in sows fed the high-protein diet. Serum interferon-gamma (tendency, P < 0.10) and interleukin (IL)-4 (P < 0.05) were greater, but IL-2 tended to be reduced (P < 0.10) in sows fed the high-protein diet. Colostrum immunoglobulin G and concentrations of fat, protein, urea nitrogen, lactose, and immunoglobulin G were greater (P < 0.001) in milk from sows fed the high-protein diet than the low-protein diet, but no differences between treatments for colostrum fat and milk somatic cell count.

Key points

• Gestating sows fed the high-protein diet had greater nitrogen intake, absorption, and retention (g/d), which confirms that reduced inclusion of soybean meal may limit nitrogen availability for protein deposition.
• Dietary protein did not influence reproductive performance. However, sows fed the low-protein diets had an increased rectal temperature and more mummified fetuses.
• High-protein diets improved immune responses and improved milk composition, with greater fat, protein, lactose, and IgG concentrations indicating improved milk quality and nutrients to be transferred to piglets.

Table 1. Effects of dietary protein on apparent total tract digestibility (ATTD) of dry matter and N balance of gestating sows¹

¹Data are least-square means of 44 and observations in the high-protein and low-protein diet, respectively.

²Calculated by dividing retained nitrogen by absorbed nitrogen and multiplying by 100 (Rojas and Stein, 2013).

Table 2. Nitrogen balance, reproductive performance, colostrum and milk compositions, and blood immune characteristics of sows in mid-gestation fed high- or low-protein diets¹

¹Least square means for nitrogen balance represent 44 and 46 observations in the high-protein low-protein diets, respectively. Least square means for all other data represent 78 and 76 observations in the high-protein low-protein diets, respectively.