Ileal Digestibility of Amino Acids in Conventional and Low-Kunitz Soybean Products Fed to Weanling Pigs*

: An experiment was conducted to determine the standardized ileal digestibility (SID) of amino acids (AA) in four sources of full-fat soybeans (FFSB) and in one source of soybean meal (SBM). The FFSB had different concentrations of trypsin inhibitor units (TIU) and included two sources of conventional FFSB, and two sources of a soybean variety that was selected for a reduced concentration of the Kunitz trypsin inhibitor. The conventional FFSB was either low temperature-processed (LT-FFSB-CV; 37.7% CP, 35.4 TIU/mg) or high temperature-processed (HT-FFSB-CV; 40.5% CP, 4.4 TIU/mg). The low-Kunitz FFSB was also either low temperature-processed (LT-FFSB-LK; 36.2% CP, 23.5 TIU/mg) or high temperature-processed HT-FFSB-LK; (38.2% CP, 4.0 TIU/mg). The SBM contained 47.5% CP and 3.20 TIU/mg. Twelve weanling barrows (initial BW: 11.1 ± 1.3 kg) were fitted with a T-cannula in the distal ileum. Pigs were allotted to a replicated 6 × 6 Latin square design with six diets and six periods per square. Five diets were prepared using each of the soybean sources as the only source of AA in the diet. An N-free diet was also included in the experiment to measure basal endogenous losses of AA. The two low temperature-processed FFSB had lower (p<0.05) AID and SID values for all indispensable AA than the two high temperature-processed FFSB and SBM. The SID values for all indispensible AA except Trp were greater (p<0.05) in LT-FFSB-LK than in LT-FFSB-CV, but the SID of AA in HT-FFSB-CV and HT-FFSB-LK were not different. The SID of AA in SBM were not different from the SID in HT-FFSB-CV and in HT-FFSB-LK. Results of this experiment show that a reduction of the TIU from 35.4 to 23.5 TIU/mg will improve the SID of AA, but this reduction is not sufficient to completely ameliorate the negative impact of trypsin inhibitors. Results also show that the SID of AA in high temperature-processed FFSB is similar to that in de-hulled SBM.


INTRODUCTION
Trypsin inhibitors are the most important antinutritional factors in raw soybeans. Kunitz and Bowman-Birk are the two major types of trypsin inhibitors in soybeans (Rackis, 1972), but the Kunitz trypsin inhibitor is of particular interest because it is heat labile, whereas the Bowman-Birk inhibitor exhibits a considerable resistance to heat treatment (Clemente et al., 2007). After isolation and characterization of the Kunitz trypsin inhibitor (Kunitz, 1947a, b), it was demonstrated that this inhibitor results in decreased protein digestibility in pigs due to a reduction in the activity of trypsin, chymotrypsin, and other pancreatic enzymes (Yen et al., 1977).
Heat treatment of soybean products inactivates the Kunitz inhibitor (Liener and Kakade, 1980) and heat treatment of soybean products is, therefore, routinely done before soybean products are used in diets fed to swine. The apparent ileal digestibility (AID) of amino acids (AA) in soybeans is also improved with heat treatment by approximately 15 percentage units (Herkelman et al., 1992), but there is no information about the effect of trypsin inhibitors on the standardized ileal digestibility (SID) of AA in soybean meal (SBM). Because of the negative impact of the Kunitz trypsin inhibitor on protein digestibility, plant breeders have tried to select varieties of soybeans with a low concentration of the Kunitz inhibitor (Clark and Hymowitz, 1972). Previous research has demonstrated that growth performance in pigs is improved if unheated varieties of low-Kunitz soybeans instead of unheated conventional soybeans are fed to pigs (Yen et al., 1974;Cook et al., 1988;Palacios et al., 2004). However, in all of these experiments, pigs fed the low-Kunitz soybeans had performance that was lower than pigs fed heat-treated soybeans. Schillinger Genetics (Des Moines, IA, USA) has recently selected a new variety of low-Kunitz soybeans, but there are no data on the nutritional quality of this variety of soybeans.
The objective of this experiment was, therefore, to test the hypothesis that low-Kunitz soybeans from Schillinger Genetics Inc. have greater AID and SID of crude protein (CP) and AA than conventional soybeans with normal concentrations of trypsin inhibitors and that heat treatment of low-Kunitz soybeans is not necessary to maximize AA digestibility.

Animals, housing, and experimental design
The experimental protocol for this experiment was reviewed and approved by the Animal Care and Use Committee at the University of Illinois. Twelve growing barrows (initial BW: 11.1±1.3 kg) were allotted to a replicated 6×6 Latin square design with 6 diets and 6 periods balanced for potential residual effects using the Balanced Latin Square Designer (Kim and Stein, 2009). Each pig was surgically equipped with a T-cannula in the distal ileum using procedures adapted from Stein et al. (1998). Pigs were housed in individual pens (1.8×2.7 m) in an environmentally controlled room. A feeder and a nipple drinker were installed in each pen.

Diets and feeding
Five sources of soybean products were used in the experiment ( Table 1). The control source was a conventional dehulled SBM containing 47.5% CP. Two Table 1. Chemical composition of soybean products produced from low temperature-processed conventional full fat soybeans (LT-FFSB-CV), low temperature-processed low-Kunitz soybeans (LT-FFSB-LK), high temperature-processed conventional soybeans (HT-FFSB-CV), high temperature-processed low-Kunitz soybeans (HT-FFSB-LK), and in conventional soybean meal (SBM), as-is basis sources of full-fat soybeans (FFSB) were also used. One of these FFSB was selected for a low concentration of Kunitz trypsin inhibitors and the other source was a conventional FFSB with normal activity of trypsin inhibitors. After isolation and characterization of the Kunitz trypsin inhibitor (Kunitz, 1947a, b), it was demonstrated that this inhibitor results in decreased protein digestion in pigs (Yen et al., 1977). Therefore, a newly developed variety with a low concentration of the Kunitz trypsin inhibitor was used. The two varieties of soybeans were grown in northeastern Indiana and were planted May 20, 2008, and harvested October 15, 2008. The growing season began very wet and cool and finished with very dry conditions. The conventional and low-Kunitz sources of soybeans were used as low temperature-processed FFSB (LT-FFSB-CV and LT-FFSB-LK) and high temperature-processed FFSB (HT-FFSB-CV and HT-FFSB-LK, respectively). The LT-FFSB-LK and LT-FFSB-CV were both dehulled and cracked through a Roskamp Double Roller Mill (Roskamp Champion, Waterloo, IA) at an ambient temperature of 21°C. The high temperature-processed FFSB were processed using the Insta-Pro model 600 autogenous extruder (Insta-Pro International, Des Moines, IA) with a 0.8 cm die operating at a rate of 182 kg per h. The extrusion temperature was 154°C for HT-FFSB-CV and 143°C for HT-FFSB-LK. The low temperature-processed soybeans and the extrudate from both high temperature processed FFSB sources were ground to 1.18 mm through a Bauer mill in the Ag Bioprocess Laboratory at the University of Illinois.
Six diets were prepared (Tables 2 and 3). Five of the diets contained one of the soybean sources and starch, sugar, and oil. The last diet was an N-free diet that was used to calculate basal endogenous losses of CP and AA. Vitamins and minerals were included in all diets to meet or exceed current requirement estimates (NRC, 1998). All diets were supplemented with 0.4% chromic oxide as an indigestible marker.
All pigs were fed at a daily level of three times the maintenance energy requirement (106 kcal of ME per kg 0.75 ; NRC, 1998). The daily allotment of feed was provided at 0700 h. Water was available at all times throughout the experiment.

Data recording and sample collection
Pig body weights were recorded at the beginning and at the end of each period and the amount of feed supplied each day was recorded. The initial four d of each period was considered an adaptation period to the diet. Ileal digesta were collected for 8 h on d 5 and 6. The cannulas were opened and a plastic bag was attached to the cannula barrel and digesta flowing into the bag were collected. Bags were removed whenever they were filled with digesta or at least once every 30 min, and stored at -20°C to prevent bacterial degradation of the AA in the digesta.

Chemical analysis
At the conclusion of the experiment, ileal samples were thawed, mixed within animal and diet, and a sub-sample was collected for chemical analysis. A sample of each diet Table 2. Ingredient composition of experimental diets containing low temperature-processed conventional soybeans (LT-FFSB-CV), low temperature-processed low-Kunitz soybeans (LT-FFSB-LK), high temperature-processed conventional soybeans (HT-FFSB-CV), high temperature-processed low-Kunitz soybeans (HT-FFSB-LK), and conventional soybean meal (SBM), as-is basis

Calculations and statistical analysis
The AID and SID of CP and AA in the five diets containing FFSB or SBM were calculated (Stein et al., 2007). Because the soybean products were the only feed ingredients contributing CP and AA in each of the diets, these digestibility values also represented the digestibility values for each of the soybean products.
The Proc UNIVARIATE procedure (SAS Institute Inc., Cary, NC) was used to identify outliers and one outlier was removed within the LT-FFSB-CV treatment. Data were analyzed using the Proc GLM procedure of SAS. The initial model included diet, period, and animal, but period and animal were not significant, and thus, removed from the final model. Orthogonal contrasts were used to determine effects of the variety of FFSB, the thermal treatment, and their interaction. Treatment means were separated using the PDIFF option with Tukey's adjustment. The mean separation output was converted to letter groupings using a SAS macro program (Saxton, 1998). The pig was the Table 3. Chemical composition of experimental diets containing low temperature-processed conventional soybeans (LT-FFSB-CV), low temperature-processed low-Kunitz soybeans (LT-FFSB-LK), high temperature-processed conventional soybeans (HT-FFSB-CV), high temperature-processed low-Kunitz soybeans (HT-FFSB-LK), and conventional soybean meal (SBM), as-is basis experimental unit for all analyses and an alpha value of 0.05 was used to assess significance among treatments.

RESULTS
The CP and AA concentrations were greater in SBM than in any of the FFSB sources (Table 1). However, the concentration of crude fat was lower in SBM than in the other soybean products, but the concentration of sucrose was not different among the soybean products. The TIU concentration in LT-FFSB-CV (35.4 TIU/mg) was greater than in LT-FFSB-LK (23.5 TIU/mg), but HT-FFSB-CV and HT-FFSB-LK had TIU concentrations of only 4.4 and 4.0 TIU/mg, and SBM contained 3.2 TIU/mg.
The AID of CP and all AA except Pro in HT-FFSB-CV, HT-FFSB-LK, and SBM was greater (p<0.05) than the AID of AA in LT-FFSB-CV and LT-FFSB-LK (Table 4). The AID of CP and all AA except Trp in LT-FFSB-CV was less (p<0.05) than in LT-FFSB-LK, but no differences in the AID of CP and AA among HT-FFSB-CV, HT-FFSB-LK, and SBM were observed. The AID for CP and all AA was greater (p<0.05) in HT-FFSB-LK and HT-FFSB-CV than in LT-FFSB-LK and in LT-FFSB-CV and the AID of all AA except Trp and Pro were greater (p<0.05) in the 2 low-Kunitz meals than in the two conventional meals.
No differences in SID for CP and AA were observed among HT-FFSB-CV, HT-FFSB-LK, and SBM, but all of these soybean products had SID values for CP and all AA except Pro that were greater (p<0.05) than the SID of CP and AA in LT-FFSB-LK and LT-FFSB-CV (Table 5). The SID of CP and all AA except Trp in LT-FFSB-LK was greater (p<0.05) than in LT-FFSB-CV. The SID for CP and all AA except Trp and Pro in HT-FFSB-LK and LT-FFSB-LK was greater (p<0.05) than in HT-FFSB-CV and LT-FFSB-CV, and the SID of CP and all AA in HT-FFSB-LK and HT-FFSB-CV were greater (p<0.05) than in LT-FFSB- Table 4. Apparent ileal digestibility (%) of crude protein (CP) and amino acids (AA) in low temperature-processed conventional soybeans (LT-FFSB-CV), low temperature-processed low-Kunitz soybeans (LT-FFSB-LK), high temperature-processed conventional soybeans (HT-FFSB-CV), high temperature-processed low-Kunitz soybeans (HT-FFSB-LK), and in soybean meal (SBM) 1

Item
Diet SEM p-values 2, 3 a-c Means within a row lacking a common superscript letter are different (p<0.05). 1 Each least squares mean represents 12 observations. 2 p-values: Overall = The p-value for the comparison of diets containing all 5 ingredients; LT vs. HT = Comparison of the two low temperature-processed full-fat meals and the two high temperature-processed meals; CV vs. LK = Comparison of the two conventional soybeans and the two low-Kunitz soybeans. 3 The interaction of the effect of processing temperature on variety of soybean were significant (p<0.05) for CP and all AA except for Trp. LK and LT-FFSB-CV.

DISCUSSION
The two major types of trypsin inhibitors in legumes, particularly soybeans, are the Kunitz and Bowman-Birk trypsin inhibitors, which make up nearly 60% of the protein in soybeans (Losso, 2008). Bowman-Birk makes up approximately 40% of the trypsin inhibitors and Kunitz the remaining 60% (Liener, 1981). Similar to Kunitz, Bowman-Birk inhibitors reduce the activity of the proteolytic enzymes secreted by the pancreas. Heat treatment inactivates the Kunitz trypsin inhibitors in soybeans, thus allowing for increased AA digestibilities and improved growth performance (Herkelman et al., 1992;Palacios et al., 2004). However, the Bowman-Birk inhibitor looses no activity if treated at a pH of two or heated to 100°C for ten minutes (Losso, 2008). The resistance of the Bowman-Birk inhibitor to highly acidic conditions and to heat treatment up to 100°C is due to the presence of disulfide bridges within the protein molecules (Losso, 2008). This structure will tightly bind up pancreatic proteases and cause severe inhibitory responses (Clemente et al., 2008). To inactivate the Bowman-Birk inhibitors, the temperature needs to exceed 100°C. During soybean extrusion, the temperature at the end of the extruder may range from approximately 143 to 166°C, which is effective in inactivating both Kunitz and Bowman-Birk inhibitors (Webster et al., 2003). The low TIU levels in both of the high temperature-processed FFSB used in the present study is, therefore, a result of the heat treatment these products had undergone. However, if the beans are processed under low temperatures, the Bowman- Table 5. Standardized ileal digestibility (%) of crude protein (CP) and amino acids (AA) in low temperature processed conventional soybeans (LT-FFSB-CV), low temperature processed low-Kunitz soybeans (LT-FFSB-LK), high temperature processed conventional soybeans (HT-FFSB-CV), high temperature processed low-Kunitz soybeans (HT-FFSB-LK), and in soybean meal (SBM)  p-values: Overall = The p-value for the comparison of diets containing all 5 ingredients; LT vs. HT = Comparison of the two low temperature-processed full-fat meals and the two high temperature-processed meals; CV vs. LK = Comparison of the two conventional soybeans and the two low-Kunitz soybeans. 4 The interaction of the effect of processing temperature on variety of soybean were significant (p<0.05) for all AA except for CP, Trp, Ala, and Pro.