Optimal lysine in diets for high-yielding lactating sows1

Abstract

The objective of the current study was to determine the optimal concentration of dietary standardized ileal digestible (SID) Lys required to maximize litter gain and minimize sow BW loss in modern high-yielding lactating sows when SID CP was kept constant across dietary treatments. A total of 396 parity 1 to 5 sows (L × Y, DanBred, Herlev, Denmark) were included in the study from day 3 after farrowing until weaning at day 26. Sows were allocated to 6 dietary treatments increasing in SID Lys concentration (6.19, 6.90, 7.63, 8.33, 9.04, and 9.76 g/kg). Diets were isoenergetic (14.04 MJ ME/kg as-fed). Litters were standardized to 14 piglets at day 3 ± 2 d postpartum. At day 3 ± 2 d and at day 26 ± 3 d in lactation, litter weight, and sow BW and back fat were registered. On a subsample of 72 parity 2 to 4 sows, litters were additionally weighed at days 10 and 17 ± 3 d, and milk and blood were sampled at day 3 ± 2 d, and 10, 17 and at 24 ± 3 d in lactation. For the 72 sows, body pools of fat and protein were also determined at days 3 ± 2 and 26 ± 3 d using the D2O dilution technique. All data were analyzed as a randomized complete block design using PROC MIXED in SAS. Furthermore, data were subjected to linear and quadratic polynomial contrasts. Variables with quadratic or linear effects or days in milk × treatment interactions were selected for analysis in PROC NLMIXED using linear broken-line models to evaluate optimal SID Lys concentrations. Only models that converged and the best fitting models were included. Average daily litter gain increased until a breakpoint at 8.11 g/kg of SID Lys (as-fed). At and above the breakpoint, multiparous and primiparous sows had litter gains of 3.36 and 2.93 kg/d, respectively. Weaning litter size (13.0 ± 0.1) was similar between the 6 dietary treatments (P = 0.28). Lactation sow BW loss was minimized to 0.17 kg/d at 9.05 g/kg of SID Lys and sow body protein loss was minimized to 0.23 kg at 9.22 g/kg of SID Lys. Linear broken-line analyses showed that for 3, 10, 17, and 24 DIM, plasma urea was minimized at 7.02, 8.10, 8.73, and 8.32 g/kg of SID Lys, respectively, and milk fat was maximized at 7.80 g/kg of SID Lys. In conclusion, in our conditions, high-yielding lactating sows required 8.11 g/kg of SID Lys to maximize litter gain and 9.05 g/kg of SID Lys to minimize sow BW loss. Based on plasma urea, the optimal dietary concentration of SID Lys was lowest in week 1, intermediate in week 2 and 4, and greatest in week 3 of lactation.

Keywords: amino acid; blood metabolite; hyper-prolific sow; litter growth; milk composition; milk yield.

Figure 1.

The effect of increasing standardized ileal digestible (SID) Lys on litter ADG based on the 390-sow data set. Data were best described by a linear broken-line model. The breakpoint is presented together with SEM and the P-value for the slope below the breakpoint is presented. Litter ADG increased until a breakpoint at 8.11 ± 0.66 g/kg of SID Lys and reached a plateau at 2.93 kg/d and 3.36 kg for primi- and multiparous sows, respectively (P < 0.05). For X_i < 8.11, ADG = 2.93_primi or 3.36_multi − 0.10 × (8.11 − X_i). The value, X_i, is the concentration of SID Lys, g/kg, for the individual sow, i. The symbols, black diamonds (♦), are the least squared means from the ANOVA test for dietary treatment 1 through 6. The vertical lines (|) are the 95% CI for the least squared means within each treatment. The fitted models from the NLmixed procedure are plotted as the given lines.

Figure 2.

Litter ADG, for the 71-sow data set, at 3 to 10, 10 to 17, and 17 to 24 days in milk (P < 0.001). Error bars indicate the SEM.

Figure 3.

The effect of increasing standardized ileal digestible (SID) Lys on (A) sow BW loss based on the 390-sow data set and (B) sow body protein loss based on the 71-sow data set. Data were best described by linear broken-line models. The breakpoints are presented together with SEM and the P-values for the slopes below the breakpoints are presented. (A) Sow BW loss decreased until a breakpoint at 9.05 ± 0.01 g/kg of SID Lys and reached a plateau at 0.17 kg/d (P < 0.001). For X_i < 9.05, sow BW loss, kg/d = 0.17 + 0.12 × (9.05 − X_i). (B) Sow body protein loss decreased until a breakpoint at 9.22 ± 0.96 g/kg of SID Lys and reached a plateau at 0.23 kg (P < 0.05). For X_i < 9.22, sow body protein loss, kg = 0.23 + 0.7 × (9.22 − X_i). Negative values indicate a gain. The value, X_i, is the concentration of SID Lys, g/kg, for the individual sow, i. The symbols, black diamonds (♦), are the least squared means from the ANOVA test for dietary treatment 1 through 6. The vertical lines (|) are the 95% CI for the least squared means within each treatment. The fitted models from the NLmixed procedure are plotted as the given lines.

Figure 4.

The effect of increasing standardized ileal digestible (SID) Lys on (A) plasma NEFA (based on log-transformed data) and (B) plasma urea nitrogen. The breakpoints are presented together with SEM and the P-values for the slopes above or below the breakpoints are presented. (A) The log-transformed data for plasma NEFA were best described by 2-slope linear broken-line models: Y_i = 5.97_DIM3 or 5.46_DIM10 or 4.77_DIM17 or 4.57_DIM24 − 0.43 × (7.50 − X_i) for X_i < 7.50 and Y_i = 5.97_DIM3 or 5.46_DIM10 or 4.77_DIM17 or 4.57_DIM24 − 0.22 × (X_i − 7.50) for X_i > 7.50. Plasma NEFA increased (P = 0.13) until a breakpoint of 7.50 ± 0.42 g/kg of SID Lys reaching 392, 236, 118, and 96 µmol/L (back-transformed estimates) at 3, 10, 17, and 24 days in milk (DIM), respectively, and thereafter, it decreased at greater SID Lys levels (P = 0.11). (B) Plasma urea nitrogen was best described by 1-slope linear broken-line models: Y_{i DIM3} = 2.34 + 0.79 × (7.02 − X_i) for X_i < 7.02, Y_{i DIM10} = 3.06 + 0.63 × (8.10 − X_i) for X_i < 8.10, Y_{i DIM17} = 3.43 + 0.62 × (8.73 − X_i) for X_i < 8.73, Y_{i DIM24} = 3.44 + 0.79 × (8.32 − X_i) for X_i < 8.32. At 3, 10, 17, and 24 DIM, PUN decreased until breakpoints of 7.02 ± 0.30 g/kg of SID Lys at 2.34 mmol/L (P = 0.10), 8.10 ± 0.38 g/kg of SID Lys at 3.06 mmol/L (P = 0.01), 8.73 ± 0.33 g/kg of SID Lys at 3.43 mmol/L (P < 0.01), and 8.32 ± 0.26 g/kg of SID Lys at 3.44 mmol/L (P < 0.01), respectively. The value, X_i, is the concentration of SID Lys, g/kg, for the individual sow, i. The least squared means from the ANOVA test are plotted as the given symbols, whereas the fitted models from the NLmixed procedure are plotted as the given lines.

Figure 5.

The effect of increasing standardized ileal digestible (SID) Lys on (A) milk fat, (B) milk DM, (C) milk energy, and (D) milk urea. The breakpoints are presented together with SEM and the P-values for the slopes above or below the breakpoints are presented. (A) Milk fat was best described by 1-slope linear broken-line models; Y_i = 8.32_DIM3 or 6.98_DIM10 or 6.93_DIM17 or 6.38_DIM24 − 0.49 × (7.80 − X_i) for X_i < 7.80. Milk fat increased until a breakpoint of 7.80 ± 0.36 g/kg of SID Lys reaching 8.32, 6.98, 6.93, and 6.38% at 3, 10, 17, and 24 DIM (P < 0.05), respectively. (B) Milk DM was best described by 1-slope linear broken-line models; Y_i = 19.3_DIM3 or 17.7_DIM10 or 17.6_DIM17 or 17.3_DIM24 − 0.49 × (7.81 − X_i) for X_i < 7.81. Milk DM increased until a breakpoint of 7.81 ± 0.51 g/kg of SID Lys reaching 19.3, 17.7, 17.6, and 17.3% at 3, 10, 17, and 24 DIM (P = 0.06), respectively. (C) Milk energy was best described by 1-slope linear broken-line models; Y_i = 5.35_DIM3 or 4.69_DIM10 or 4.65_DIM17 or 4.52_DIM24 − 0.20 × (7.66 − X_i) for X_i < 7.66. Milk energy increased until a breakpoint of 7.66 ± 0.29 g/kg of SID Lys reaching 5.35, 4.69, 4.65, and 4.52 MJ/kg at 3, 10, 17, and 24 DIM (P < 0.01), respectively. (D) Milk urea was best described by 2-slope linear broken-line models; Y_i = 28.6_DIM3 or 29.4_DIM10 or 32.7_DIM17 or 39.5_DIM24 + 10.8 × (7.33 − X_i) for X_i < 7.33 and + 1.22 × (X_i − 7.33) for X_i > 7.33. Milk urea decreased (P < 0.05) until a breakpoint of 7.33 ± 0.22 g/kg of SID Lys reaching 28.6, 29.4, 32.7, and 39.5 mg/dL at 3, 10, 17, and 24 DIM, respectively, and thereafter, it increased slightly at greater SID Lys levels (P = 0.23). The value, X_i, is the concentration of SID Lys, g/kg, for the individual sow, i. The least squared means from the ANOVA test are plotted as the given symbols, whereas the fitted models from the NLmixed procedure are plotted as the given lines.