Isoleucine requirement of pregnant sows 1

Transcription

1 Published November 25, 2014 Isoleucine requirement of pregnant sows 1 D. J. Franco,* J. K. Josephson,* S. Moehn,* P. B. Pencharz, and R. O. Ball* 2 *Department of Agriculture, Food and Nutritional Sciences, University of Alberta, Edmonton, AB T6G 2P5, Canada; and Research Institute, Hospital for Sick Children, Toronto, ON M5G 1X8, Canada ABSTRACT: The objective of this study was to determine the Ile requirement in early (d 39 to 61) and late (d 89 to 109) pregnancy using the indicator AA oxidation method. The same 7 Large White Landrace sows in their fourth parity were used in early and late pregnancy. Each sow received 6 diets based on corn, corn starch, and sugar in both early and late pregnancy at constant feed allowances (2.5 kg/d). Diets provided Ile at 20, 40, 60, 80, 100, and 120% of the Ile requirement (6.2 g/d based on the 1998 NRC) in early and 60, 80, 100, 140, 160, and 180% in late pregnancy. After determination of 13 C background in expired CO 2 and plasma free Phe for 1.5 h when confined in respiration chambers, sows were fed the tracer, L[1-13 C]Phe, a rate of 2.0 mg/(kg BW h) over 4 h divided into eight 30-min meals. Expired CO 2 and plasma free Phe were analyzed for 13 C enrichment above background. Requirements were determined as the breakpoint in 2-phase nonlinear models. Sow BW was kg in early and kg in late pregnancy. Daily gain of the 6 sows was similar in early (344 g/d) and late pregnancy (543 g/d). During pregnancy, sow maternal gain was 19.1 ± 4.4 kg and litters of 17.7 ± 0.8 piglets weighed 22.6 ± 0.9 kg at birth. The Ile requirement was 3.6 ± 1.2 g/d (P = 0.001) in early pregnancy with a Phe retention ( 0.59 g/d) and energy retention ( 0.31 MJ/d) that were not different from 0. This indicates that the fourth parity sows had requirements close to maintenance in early pregnancy. The Ile requirement in late pregnancy was 9.7 ± 1.9 g/d (P = 0.001) when sows retained 3.30 g/d of Phe and 1.45 MJ/d of energy. The greater Ile requirement in late pregnancy was probably caused by the increased conceptus growth after d 70 of pregnancy. Phenylalanine flux, oxidation, and nonoxidative disposal increased (P < 0.1) from early to late pregnancy, but body protein breakdown did not. Phenylalanine oxidation, nonoxidative disposal, and retention increased (P < 0.01) with increasing Ile intake in early pregnancy but were not affected by Ile intake in late pregnancy. Body protein breakdown did not respond to Ile intake in early or late pregnancy. Although energy retention was similar in early and late pregnancy, the respiratory quotient decreased (P = 0.047) from early (1.05) to late pregnancy (0.98), indicating lipid mobilization in late pregnancy when Ile was at or above the requirement. The results of this study show that the Ile requirement of sows increases from early to late pregnancy. Key words: gestation, L-isoleucine, requirement, sow 2013 American Society of Animal Science. All rights reserved. J. Anim. Sci : doi: /jas INTRODUCTION The assessment of pregnant sow AA requirements changed in that AA requirements were predicted based on models rather than derived from empirical studies. 1 Supported by funds from Alberta Pork (Edmonton, AB, Canada), Alberta Agricultural Research Institute (Edmonton, AB, Canada), Ontario Pork (Guelph, ON, Canada), Ajinomoto Heartland LLC (Raleigh, NC), and Advancing Canadian Agriculture and Agri-Food (Ottawa, ON, Canada). Ajinomoto Heartland LLC (Raleigh, NC) provided crystalline AA. 2 Corresponding author: ron.ball@ualberta.ca Received December 12, Accepted April 12, Models from France (Dourmad et al., 2008), Germany (GfE, 2008), and the United States (Kim et al., 2009; NRC, 2012) are typically based on the growth of body components (i.e., maternal body, placenta, fetuses, and mammary growth). These models contain estimates for the AA composition of protein deposition in these body components and for maintenance, and an estimate of the efficiency of AA retention to calculate AA requirements. Among these models, there is agreement that the requirements for AA and energy increase in late pregnancy. This trend is supported by recent empirical studies that showed that the requirements for Lys (Srichana, 2006; Samuel et al., 2012) and Thr

2 3860 Franco et al. (Levesque et al., 2011) were greater in late compared with early pregnancy. In addition, the GfE (2008) and the NRC (2012) models indicate that the AA requirements of pregnant sows may decrease when sows approach maturity because maternal growth decreases as the sows approach mature body size. Empirical determinations of AA requirements are most frequently performed in gilts and young sows so that data for older sows are scarce. Furthermore, there are few studies on pregnant sow AA requirements in recent years despite a large increase in sow performance (CCSI, 2010). For Ile, the fourth limiting AA in cornsoybean meal diets (NRC 1998), there has been only 1 determination of requirements (Speer et al., 1990) that was performed in pregnant gilts. Because of the lack of empirical information on AA requirements in modern genotypes, especially in older sows, the experiments reported herein were conducted to determine the Ile requirement in early and late pregnancy in sows approaching adult age and size, and to compare determined with predicted requirements based on the modeling approach. Materials and Methods Animal care and experimental procedures were the same for each experiment and were approved by the University of Alberta Animal Care and Use Committee for Livestock (Edmonton, AB, Canada). The study was conducted to determine the response of indicator AA oxidation in early (d 39 to 61) and late pregnancy (d 89 to 109) to increasing amounts of dietary Ile. Each of 7 Large White Landrace sows pregnant in their fourth parity were fed 6 diets of increasing dietary Ile in random order in both early and late pregnancy. Dietary Ile contents ranged from deficient to excess compared with the estimated (NRC 1998) requirement (6.2 g/d of total Ile) for sows of similar BW at breeding, anticipated maternal growth (40 kg) and litter size (12 piglets). The sows had been left to return to service after weaning to synchronize pregnancy with the availability of experimental equipment. Experimental Diets The diets were based on corn, cornstarch, and sugar (Table 1). Corn was included in the low-ile diet to meet the least amount of dietary Ile (i.e., 20% of requirement in early and 60% of requirement in late pregnancy). In the high-ile diets (Table 1), L-Ile was included at the expense of corn starch, sugar, and Gly to obtain Ile contents of 120% of requirement in early and 180% of requirement in late pregnancy. The low- and high- Ile diets were cross-mixed within stage of pregnancy to Table 1. Ingredient composition of the low-ile and high- Ile diets as fed to determine the Ile requirement of sows in early and late pregnancy Early pregnancy Late pregnancy Item Ile intake, 1 %: Ingredient, % Corn Corn starch Sugar (sucrose) Powdered cellulose Canola oil Celite L-His L-Ile L-Leu L-Lys HCl L-Cys DL-Met L-Phe L-Tyr L-Thr L-Trp L-Val Gly Micro mineral premix Vitamin premix Choline choride Dicalcium phosphate Limestone Sodium bicarbonate Salt Potassium chloride Calculated energy and nutrient contents Ile, g/kg CP, % ME, MJ/kg NRC, Contained per kilogram of diet: 30 mg of Cu as copper sulfate; 45 mg of Fe as ferrous sulfate; 15 mg of Mn as manganous oxide; 0.18 mg of Se as sodium selenite; 75 mg of Zn as zinc oxide; and 0.3 mg of I as potassium iodide (DSM Nutritional Products, Hay River, AB, Canada). 3 Contained per kilogram diet: 9600 IU of vitamin A; 960 IU of vitamin D 3 ; 64 IU of Vitamin E; 5.12 mg of vitamin K; 3.2 mg of thiamin; 6.4 mg of riboflavin; 1.92 mg of pyridoxine; 19.2 µg of vitamin B 12 ; 48 mg of niacin; 19.2 mg of pantothenic acid; 3.2 mg of folic acid; 0.32 g of biotin (DSM Nutritional Products). obtain diets with Ile levels of 40, 60, 80, and 100% of the calculated requirement in early and of 80, 100, 140, and 160% in late pregnancy. All other indispensable AA were provided at a minimum of 120% in early and 180% in late pregnancy of their respective NRC (1998) requirement. The vitamin and mineral premix was added to meet or exceed NRC (1998) requirements. The inclusion of NaCl, NaHCO 3, and KCl, differed among diets to balance dietary electrolytes. Analyzed diet contents for GE, CP, and AA are shown in Table 2.

3 Isoleucine requirement of pregnant sows 3861 Table 2. Analyzed nutrient and energy contents (as fed) of the low Ile and high Ile diets to determine the Ile requirement of sows in early and late pregnancy 1 Early pregnancy Late pregnancy Item Ile intake, 2 %: CP, % GE, MJ/kg Indispensable AA, % Arg His Ile Leu Lys Met Phe Thr Trp Val Dispensable AA, % Ala Asp Cys Glu Gly Pro Ser Tyr 0.15 ` Amino acid analysis was performed by the Experiment Station Chemistry Laboratory, University of Missouri (Columbia, MO). 2 NRC, Dietary Phe and Tyr concentrations were kept constant in all diets. The sows were offered their daily feed allowance in 2 equal daily feedings except on respiration days when one-half the daily allowance was divided into twelve 30-min portions based on Moehn et al. (2004a). Feed allowance was set to achieve the recommended daily energy intake based on BW and backfat at breeding according to Aherne and Foxcroft (2000) and kept constant throughout gestation. All animals were individually housed. The high Ile diet was fed for 7 d before the change to the assigned diet for the first respiration day. A minimum of 3 d of adaptation was used for each successive diet based on the reports that this was a more-than-adequate adaptation period for respiration experiments (Moehn et al., 2004b; Elango et al., 2009). Administration of Labeled Amino Acids In both experiments, sows received an oral dose of 2.0 mg/(kg of BW h) of L-[1-13 C]Phe (99% enrichment; Sigma Aldrich, Mississauga, ON, Canada) for 4 h divided into eight 30-min feedings. A priming dose equal to 1.75 times the hourly dose was given along with the first 30-min dose. Sample Collection Sows were fitted with a surgically implanted coated catheter (CBAS 100 cm catheter; Instech Solomon, Plymouth Meeting, PA) and vascular access titanium injection port (TiSoloPort MAX; Instech Solomon; Swindle et al., 2005). The vascular port was accessed before sows were placed in respiration chambers, and an extension set was externalized from the chamber to facilitate blood sampling. Sows were placed in the respiration chambers at 1600 h the day before the beginning of the collection period to allow the air in the chamber to equilibrate with the ventilating air stream. Design and operation of the respiration chambers was described in detail by Levesque et al. (2011). Apparent oxidation of Ile was measured in an aliquot of CO 2 from expired air after administration of L-[1-13 C]Phe. Expired air was collected in 30-min intervals into 11-mL of 1 N NaOH solution. Background 13 CO 2 enrichment was measured for three 30-min periods before administration of the isotope. Blood samples (5 ml) were collected in heparinized Vacutainer tubes (Fisher Scientific, Mississauga, ON, Canada) at 30-min intervals immediately before feeding during the 5.5-h collection period. Blood samples were centrifuged at 1500 g for 15 min at room temperature for separation of plasma. Analyses Expired air samples were analyzed for 13 CO 2 after the procedure reported by El-Khoury et al. (1994). Expired CO 2, which was trapped in 1 N NaOH solution in the form of Na 2 CO 3, was reacted with H 3 PO 4 to release the free CO 2 gas. The 13 CO 2 enrichment in the gas was then measured by a continuous flow, dualinlet isotope ratio mass spectrometer (CF-IRMS 20/20 Isotope Analyzer; PDZ Europa Ltd., Cheshire, UK; Di Buono et al., 2001). Each set of 8 samples (2 baseline and 6 after isotope dosing) was separated by reference samples (5% CO 2 ), which were previously calibrated to an international reference standard (NBS-20; National Institute for Standards and Technology, Gaithersburg, MD). The 13 C enrichment in plasma free Phe was analyzed using a triple quadrupole mass analyzer (API4000; Applied Biosystems/MDS SCIEX, Concord, ON, Canada) coupled to a HPLC system (Aligent 1100 HPLC; Aligent, Mississauga, ON, Canada) as previously described by Chapman et al. (2009). Reverse phase HPLC (Waters, Millipore, Mississauga, ON, Canada) with the use of phenylisothiocyanate derivatives was used to measure plasma AA (Bidlingmeyer et al., 1984). Gross energy of feed samples was determined using an adiabatic bomb calorimeter (Leco AC-300; LECO Corp., St. Joseph, MI). Nitrogen contents of the samples were determined by complete combustion and

4 3862 Franco et al. measurement of NO gas contents by infrared radiation (LECO Corporation). Amino acids in feed were analyzed (University of Missouri Experiment Station, Columbia, MO) using ion-exchange chromatography coupled with postcolumn ninhydrin derivatization. Calculation of Results Expired air 13 CO 2 enrichment was calculated as the difference in isotopic abundance at plateau and natural (baseline) abundance and was expressed as atom percentage excess. Plateau 13 CO 2 enrichment value for each sow/diet combination was determined as the data points where the linear regression of enrichment on collection period was not different from 0. The L-[1-13 C]Phe oxidation rates were expressed as a percentage of the infused dosage and calculated according to the equation: Phe oxidation = F 13 CO 2 /Phe inf, where F 13 CO 2 = the rate of 13 CO 2 released by Phe tracer oxidation (mmol 13 CO 2 /30 min), and Phe inf = rate of L-[1-13 C] Phe tracer administered (mmol 13 C-Phe/30 min). The F 13 CO 2 was calculated according to the equation (Matthews et al., 1980): F 13 CO 2 = (FCO 2 ECO )/( ), where FCO 2 = rate of CO 2 production (ml/min) and ECO 2 = 13 CO 2 enrichment in expired air at isotopic steady state (atom percentage excess). The constants 30 min and 42.3 mmol/ml convert FCO 2 to mmol/30 min, and the factor 100 changes atom percentage excess to a fraction. The constant 0.82 accounts for 13 CO 2 retained in the body because of bicarbonate fixation (Moehn et al., 2004a). Plasma Phe flux was determined at isotopic steady state using the equation of Matthews et al. (1980): Q = i (E i /E p 1), where E i is the enrichment of L-[1-13 C] Phe infused, E p is the 13 C enrichment of plasma Phe above baseline at isotopic plateau, and i is the rate of L-[1-13 C] Phe infused. Phenylalanine used for nonoxidative disposal (protein synthesis; PS) and derived from protein (B) was calculated after Matthews et al. (1980): Q = PS + PO = B + PI, where PO is Phe oxidation (g/d) and PI is the sum of Phe intake (g/d) plus the L-[1-13 C]Phe administered during the study. Phenylalanine retention was calculated as PI PO. The formula by Brouwer (1965) was used to calculate heat production from indirect calorimetry: heat production = 1.44 [(16.18 VO 2 ) + (5.02 VCO 2 ) (2.17 VCH 4 )], where VO 2, VCO 2, and VCH 4 are the volumes (L) of O 2, CO 2, and CH 4 produced, respectively. The formula was abbreviated by omitting the urinary N. The effect of ignoring the urinary N (i.e., protein metabolism) is 1% for every 12.3% of the total energy that was derived from protein (Weir, 1949). Statistical Analysis Daily feed intake was not equal among sows. Therefore, daily Ile intake was expressed as grams Ile/day rather than percentage dietary Ile. Data were tested for outliers using the REG procedure (SAS Inst. Inc., Cary, NC). Data used to determine Ile requirement in early and late pregnancy (as % of dose) were analyzed separately. Correlation analysis, using the CORR procedure of SAS, was used to identify covariates affecting the dependent variable response (e.g., L-[1-13 C]Phe oxidation) irrespective of the effect of dietary Ile intake. Sow BW, daily gain in early and late pregnancy, parity, litter weight, litter size, and room temperature were tested as potential covariates. Stepwise regression was used to ensure all covariates identified entered the model at P < Mixed models with linear and quadratic effects of Ile and significant covariates contained sow, block, and the diet fed in the previous block as random variables. These models were used to create predicted values for breakpoint analysis. The NLIN procedure of SAS was used to determine the broken-line regression model of best fit. The confidence interval of the breakpoint was estimated according to Seber (1977). Quantitative variables of Phe kinetics and energy metabolism in early and late pregnancy were analyzed using the mixed model procedure with the main effects of Ile intake. The effects of block and stage of gestation on variables of Phe kinetics and energy metabolism were tested by least square means comparison in mixed models. For all analyses, P 0.05 and P 0.10 were considered as evidence of significance and a trend, respectively. Results The 7 sows weighed an average of kg at breeding and had a backfat depth of 18.0 ± 0.6 mm. Sows consumed 2497 ± 45 g/d of the test feed. Body weight in early pregnancy (246.3 ± 2.1 kg) was less (P = 0.003) than in late pregnancy (272.1 ± 2.1 kg). Daily BW gain in early (344 ± 75 g/d) and late pregnancy (543 ± 105 g/d) were not different. The sows had a maternal BW gain of 19.1 ± 4.4 kg in pregnancy to reach a postfarrowing BW of ± 5.5 kg. The mean litter size was 17.7 ± 0.8 total born and 17.1 ± 0.7 piglets born alive, weighing 22.6 ± 0.9 kg at birth. Indicator AA oxidation in early pregnancy responded linearly (P = 0.001) and quadratically (P = 0.001) to increasing Ile intake. Indicator AA oxidation in early pregnancy decreased with increasing growth rate (P = 0.011) and increased with increasing background 13 CO 2 (P = 0.001). In late pregnancy, the indicator AA oxidation response to increasing Ile was linear (P = 0.053) and quadratic (P = 0.065). Indicator oxidation increased with increasing pregnancy BW gain

5 Isoleucine requirement of pregnant sows 3863 Table 3. Phenylalanine kinetics and energy metabolism in early and late pregnancy Figure 1. The requirements (breakpoints) were based on analysis of values predicted from measured indicator AA oxidation. Vertical error bars indicate the SE of Phe oxidation within Ile levels; horizontal error bars the SE of Ile intake within Ile levels. Early pregnancy: Phe oxidation = 24.5 ± ± 0.21 Ile intake for Ile intake of <3.6 g/d, and Phe oxidation = 17.26% of dose for Ile intake of > 3.6 g/d (P = 0.001). Late pregnancy: Phe oxidation = 20.4 ± ± 0.11 Ile intake, for Ile intake of < 9.7 g/d and Phe oxidation = 16.71% of dose for Ile intake of > 9.7 g/d (P = 0.008). (P = 0.025) and decreased with increasing sow backfat (P = 0.002). Breakpoint analysis of indicator oxidation indicated that the dietary Ile requirement of pregnant sows was less (P = 0.007) in early pregnancy at 3.6 ± 1.2 g/d (P = 0.001) than in late pregnancy at 9.7 ± 1.9 g/d (P = 0.001; Fig. 1). All variables of quantitative Phe kinetics (Table 3) increased (P < 0.1) from early to late pregnancy except for Phe derived from body protein breakdown, which was similar for early and late pregnancy. The partitioning of Phe flux to protein synthesis or oxidation was similar in early and late pregnancy. However, Phe intake represented a greater percentage of flux in late pregnancy than in early pregnancy (P = 0.001), whereas Phe from body protein breakdown contributed less to Phe flux in late than in early pregnancy (P = 0.001). Heat production and energy retention were not affected by stage of pregnancy but the respiratory quotient was less (P = 0.001) in late than in early pregnancy. In early pregnancy, Phe oxidized and retained responded quadratically (P < 0.05) to increasing Ile intake (Table 4), as did the fraction of flux oxidized and used for protein synthesis (P < 0.1). Nonoxidative Phe disposal increased (P = 0.017) with increasing Phe intake. Retained Phe (P = 0.067) tended to increase with increasing BW, but retained energy and the respiratory quotient decreased (P < 0.10) with increasing BW. Heat production in early pregnancy increased (P = 0.008) with increasing feed intake. Indicator oxidation was least in the first block in early pregnancy, and increased (P = 0.005) as pregnancy progressed. Phenyalanine retention decreased from Blocks 1 to 6 (P = 0.005). Item Early pregnancy Late pregnancy SEM P-value n n Ile intake, g/d Phe flux, g/d Phe oxidation, g/d Phe for protein synthesis, g/d Phe intake, g/d Phe from body protein breakdown, g/d Phe retention, g/d Phe oxidation, % of flux Phe for protein synthesis, % of flux Phe intake, % of flux Phe from body protein breakdown, % of flux Phe retention, % of flux Heat production, MJ/d Energy retention, MJ/d Respiratory quotient Observations based on plasma free Phe atom percentage excess. Missing values were due to catheter malfunction. Heat production decreased, and energy retention and respiratory quotient increased from Blocks 1 to 6 in early pregnancy (P < 0.05). In late pregnancy (Table 5), neither Phe kinetics nor energy metabolism responded to Ile intake. Increasing feed intake increased oxidation and decreased Phe retention (P < 0.05). Increasing litter size decreased Phe oxidation and increased Phe retention (P = 0.001). Phenylalanine flux and Phe derived from body protein breakdown in late pregnancy increased from Blocks 1 to 6 (P = 0.008). Discussion The Ile requirement of gestating fourth parity sows increased from 3.6 g/d in early to 9.7 g/d in late pregnancy. The increase in requirement from early to late pregnancy is in agreement with recent recommendations (GfE, 2008; Kim et al., 2009; NRC, 2012) and with the results we obtained for Thr (Levesque et al., 2011), Lys (Samuel et al., 2012), and Trp (Moehn et al., 2012b). However, the magnitude of changes in Ile requirements in gestation differs between recommendations. In the current study, the Ile requirement in late pregnancy was more than double the requirement found in early pregnancy. The key for the large discrepancy between early and late pregnancy requirements compared with previous studies (Samuel et al., 2012) is the age of sows. The sows in the current experiment (fourth parity) approached adult age and had considerably less maternal BW gain (19.1 kg) than that

6 3864 Franco et al. Table 4. Phenylalanine kinetics and energy metabolism of sows fed graded levels of Ile in early gestation Ile intake, 1 % P-value Item SEM Linear Quadratic n n Ile intake, g/d Phe flux, g/d Phe oxidation, g/d Phe for protein synthesis, g/d Phe intake, 3 g/d Phe from body protein breakdown, g/d Phe retention, 4 g/d Phe oxidation, % of flux Phe for protein synthesis, % of flux Phe intake, % of flux Phe from body protein breakdown, % of flux Phe retention, % of flux Heat production, 3 MJ/d Energy retention, 5 MJ/d Respiratory quotient NRC, Observations based on plasma free Phe atom percentage excess. Missing values were due to catheter malfunction. 3 Increased with feed intake, P < Increased with BW, P < Decreased with increasing BW, P < Table 5. Phenylalanine kinetics and energy metabolism of sows fed graded levels of Ile in late gestation Ile intake, 1 % P-value, Item SEM linear n n Ile intake, g/d Phe flux, 3 g/d Phe oxidation, 4,5 g/d Phe for protein synthesis, g/d Phe intake, 3,4 g/d Phe from body protein breakdown, 3 g/d Phe retention, 6,7 g/d Phe oxidation, % of flux Phe for protein synthesis, % of flux Phe intake, % of flux Phe from body protein breakdown, % of flux Phe retention, % of flux Heat production, MJ/d Energy retention, MJ/d Respiratory quotient NRC, Observations based on plasma free Phe atom percentage excess. Missing values were due to catheter malfunction. 3 Increased with increasing BW, P < Increased with increasing feed intake, P < Decreased with increasing litter size, P = Decreased with increasing feed intake, P = Increased with increasing litter size, P =

7 Isoleucine requirement of pregnant sows 3865 observed in our previous studies (>40 kg; Samuel et al., 2012). In fact, the Phe retention in early pregnancy was not different from 0 but was greater than 0 in late pregnancy. It appears, the Ile requirement in early pregnancy was close to the Ile maintenance requirement (Moehn et al., 2012a), whereas the requirement in late pregnancy was mainly driven by the requirement for maintenance plus growth of conceptus. This clearly indicates that stage of pregnancy and parity affects sow AA requirements and deserves consideration in gestation diet formulation. The observed Phe retention of 3.6 g/d in late pregnancy for sows with Ile intake at or above requirement, approximately 100 g/d protein deposition, was greater than the protein deposition predicted by the NRC (2012) model of 72.6 g/d. Consequently, the determined Ile requirement (9.7 g/d) was greater than predicted by NRC (2012) at 6.3 g/d. In early pregnancy, the observed Phe retention of close to 0 g/d was much less than the predicted protein deposition of 38.1 g/d. Increased Phe retention was only observed in the first block in the current experiment, around d 39 of pregnancy, after missing 1 estrous cycle. Consequently, the time-dependent protein deposition was observed to a much lesser degree that predicted by NRC (2012). Similar to Samuel et al. (2012), Phe kinetics were not affected by Ile level. In agreement with Samuel et al. (2012) and Levesque et al. (2011), Phe oxidation decreased and Phe retention increased with increasing Ile level. The partitioning of Phe flux with increasing Ile level followed the pattern described by Samuel et al. (2012). Phenylalanine derived from intake and body protein breakdown were not affected, whereas Phe oxidation decreased and Phe used for protein synthesis and Phe retention both increased, although this relationship was statistically significant only in early pregnancy. It appears that increasing intake of limiting AA in pregnant sows had no effect on body protein breakdown but increased Phe retention by reducing Phe oxidation and increasing protein synthesis. This is in contrast with growing pigs, where Fuller et al. (1987) found constant flux and synthesis but reduced body protein breakdown when Lys was added to a Lysdeficient diet. The partition of Phe flux in the current experiment was similar to that in young sows (S. Moehn, unpublished data) in that the proportion of Phe intake increased, whereas Phe derived from body protein breakdown decreased from early to late pregnancy, resulting in a greater proportion of Phe flux being retained. However, although fractions of Phe flux oxidized and used for protein synthesis were similar in early and late pregnancy in the current experiment, they decreased and increased from early to late pregnancy, respectively, in the studies by Samuel et al. (2012). The younger sows used by Samuel et al. (2012) were in positive Phe balance in early pregnancy, whereas the fourth parity sows in the current experiment were at Phe equilibrium in early pregnancy. In all these experiments, Phe retention increased from early to late pregnancy but was elicited by decreased body protein breakdown in younger sows (Samuel et al., 2012) vs. increased protein synthesis in the current experiment with adult sows. This difference in response may be caused by a shift from predominantly maternal protein gain in early pregnancy to predominantly fetal protein gain in late pregnancy in younger sows vs. a shift from protein maintenance in early pregnancy to protein growth in late pregnancy in older, adult sows. Because of unchanged protein breakdown with increased intake, Phe flux increased from early to late pregnancy in the current experiment. In previous experiments, the reduced protein breakdown with marginal change in Phe intake (Samuel et al., 2012) led to a decrease of Phe flux in late vs. early pregnancy. This indicates that changes in quantitative Phe kinetics depend on diet formulation (i.e., AA content), and that adult sows may use different mechanisms to achieve increased protein deposition than younger, growing animals (Fuller et al., 1987). The age of sows may also explain the different responses in energy metabolism. Sows in the current experiment were in marginally negative energy balance in early and late pregnancy, whereas Samuel et al. (2012) reported positive energy balances that were greater in early than in late pregnancy. Although the energy balance in the current experiment was not different between early and late pregnancy, the respiratory quotient decreased from early to late pregnancy. Given the marginal energy intake in the current experiment, the sows may have had to mobilize more lipid to support the Phe retention that increased in late pregnancy. This conclusion is supported by Noblet et al. (1990), who found that older sows given a constant feed allowance tend to gain primarily lean tissue over gestation and then mobilize body fat later in pregnancy to provide energy to support conceptus growth. Therefore, feed allowances should be reevaluated not only for younger but also for adult sows. In conclusion, the Ile requirement of adult sows increased from early to late pregnancy, indicating a switch from near maintenance metabolism in early to a late pregnancy metabolism similar to younger sows. The marginal energy intake of sows in this experiment led to lipid mobilization in late pregnancy, especially at Ile levels at or above the Ile requirement, indicating that feed allowances need to be increased in late pregnancy in older sows to a similar degree as in younger sows.

8 3866 Franco et al. LITERATURE CITED Aherne, F., and G. Foxcroft Management of the gilt and first parity sow: Part V. Nutritional management in gestation and lactation. In: VII Simposio Internacional de Reproducao e Inseminacao Artificial em Suinos, Foz do Iguaçu, Brazil. p Bidlingmeyer, B. A., S. A. Cohen, and T. L. Tarvis Rapid analysis of amino acids using precolumn derivitization. J. Chromatogr. A 336: Brouwer, E Report of sub-committee on constants and factors. In: K. L. Blaxter, editor, Proc. 3rd Symp. Energy Metab. 11th ed. EAAP, Academic Press, New York. p CCSI Canadian Centre for Swine Improvement Annual Report 2009/2010. Accessed 14 May, annual/ann2010.pdf. Chapman, K. P., G. Courtney-Martin, A. M. Moore, R. O. Ball, and P. B. Pencharz Threonine requirement of parenterally fed postsurgical human neonates. Am. J. Clin. Nutr. 89: Di Buono, M., L. J. Wykes, R. O. Ball, and P. B. Pencharz Total sulfur amino acid requirement in young men as determined by indicator amino acid oxidation with L-[1-C-13] phenylalanine. Am. J. Clin. Nutr. 74: Dourmad, J.-Y., M. Étienne, A. Valancogne, S. Dubois, J. van Milgen, amd J. Noblet InraPorc: A model and decision support tool for the nutrition of sows. Anim. Feed Sci. Technol. 143: Elango, R., M. A. Humayun, R. O. Ball, and P. B. Pencharz Indicator amino acid oxidation is not affected by period of adaptation to a wide range of lysine intake in healthy young men. J. Nutr. 139: El-Khoury, A. E., N. K. Fukagawa, M. Sánchez, R. H. Tsay, R. E. Gleason, T. E. Chapman, and V. R. Young Validation of the tracer-balance concept with reference to leucine: 24-h intravenous tracer studies with L-[1-13 C]leucine and [ 15 N- 15 N] urea. J. Nutr. 59: Fuller, M. F., P. J. Reeds, A. Cadenhead, B. Seve, and T. Preston Effects of the amount and quality of dietary protein on nitrogen metabolism and protein turnover of pigs. Br. J. Nutr. 58: GfE Recommendations for the supply of energy and nutrients to pigs. DLG Verlags GmbH, Frankfurt, Germany. Kim, S. W., W. L. Hurley, G. Wu, and F. Ji Ideal amino acid balance for sows during gestation and lactation. J. Anim. Sci. 87:E123 E132. Levesque, C. L., S. Moehn, P. B. Pencharz, and R. O. Ball The threonine requirement of sows increases in late gestation. J. Anim. Sci. 89: Matthews, D. E., K. J. Motil, D. K. Rohrbaugh, J. F. Burke, V. R. Young, and D. M. Bier Measurement of leucine metabolism in man from a primed, continuous infusion of L-[1-13 C]leucine. Am. J. Physiol. 238:E473 E479. Moehn, S., R. F. Bertolo, P. B. Pencharz, and R. O. Ball. 2004a. Pattern of carbon dioxide production and retention is similar in adult pigs fed hourly, but not when fed a single meal. BMC Physiol. 4: Moehn, S., R. F. Bertolo, P. B. Pencharz, and R. O. Ball. 2004b. Indicator amino acid oxidation responds rapidly to changes in lysine or protein intake: Studies in growing and adult pigs. J. Nutr. 134: Moehn, S., D. Franco, J. K. Josephson, P. B. Pencharz, and R. O. Ball. 2012a. Isoleucine requirement for maintenance in sows. J. Anim. Sci. 90(Suppl. 2):61 (Abstr.). Moehn, S., D. Franco, J. K. Josephson, P. B. Pencharz, and R. O. Ball. 2012b. Tryptophan requirement of gestating sows. J. Anim. Sci. 90(Suppl. 2):61 (Abstr.). Noblet, J., J. Y. Dourmand, and M. Etienne Energy utilization in pregnant and lactating sows: Modeling of energy requirements. J. Anim. Sci. 68: NRC Nutrient requirements of swine. 10th rev. ed. Natl. Acad. Press, Washington, DC. NRC Nutrient requirements of swine. 11th rev. ed. Natl. Acad. Press, Washington, DC. Samuel, R. S., S. Moehn, P. B. Pencharz, and R. O. Ball The dietary lysine requirement of sows increases in late-gestation. J. Anim. Sci. 90: Seber, G. A. F Linear regression analysis. John Wiley & Sons, New York, NY. Speer, V. C., D. L. Kile, and J. C. Trew Estimation of the isoleucine and aromatic amino acid requirements of pregnant swine. J. Anim. Sci. 68: Srichana, P Amino acid nutrition in gestating and lactating sows. PhD Diss. Univ. of Missouri, Columbia, MO. Swindle, M. M., T. Nolan, A. Jacobson, P. Wolf, M. Dalton, and A. C. Smith Vascular access port (VAP) usage in large animal species. Contemp. Top. Lab. Anim. Sci. 44:7 17. Weir, J. B New methods for calculating metabolic rate with special reference to protein metabolism. J. Physiol. 109:1 9.