Effect of Heating on Nutritional Quality of Conventional and Kunitz Trypsin Inhibitor-Free Soybeans J. C. ANDERSON-HAFERMANN, Y. ZHANG, and C. M. PARSONS Department of Animal Sciences, University of Illinois, Urbana, Illinois 61801 T. HYMOWITZ Department of Agronomy, University of Illinois, Urbana, Illinois 61801
1992 Poultry Science 71:1700-1709
that the nutritional value of raw Kunitz trypsin inhibitor-free soybeans (KFSB) was A large amount of research has been substantially higher than that of raw conducted to evaluate the detrimental conventional soybeans (CSB) for chicks effects on animal performance caused by and laying hens but was less than that of antinutritional factors in raw soybeans (Friedman et ah, 1991). Kunitz trypsin commercial soybean meal (SBM). The inhibitor is one of the major antinutritional KFSB is a soybean variant that is isogenic factors in soybeans (Rackis, 1965). Han et to the commercial Williams 82 cultivar al. (1991) and Zhang et al. (1991) reported except that it lacks a functional Kunitz trypsin inhibitor allele (Bernard and Hymowitz, 1986). Friedman et al. (1991) recently showed that steam heating of Received for publication February 24, 1992. KFSB increased its protein quality for rats. Accepted for publication June 5, 1992. These results suggest that some heating of J To whom correspondence should be addressed: 322 Mumford Hall, 1301 West Gregory Drive, Urbana, KFSB is probably necessary to obtain maximal quality for poultry. However, IL 61801. INTRODUCTION
1700
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
ABSTRACT Five 10-day chick growth experiments and an amino acid digestibility assay were conducted to assess the effect of steam heating on in vivo protein quality of raw, full-fat Kunitz trypsin inhibitor-free soybeans (KFSB) compared with raw, conventional full-fat soybeans (CSB). Protein solubility in .2% KOH was also evaluated as an in vitro test of in vivo protein quality for underprocessed CSB. The CSB and KFSB were autoclaved for 0,3,6, 9,12,15,18, or 21 min at 121 C and 124 kPa. The soybeans were then fed to 8- or 9-day-old chicks as the sole source of protein in dextrose and soybean diets containing 23% protein. Growth performance of chicks fed raw KFSB was superior to that of chicks fed raw CSB. Growth performance of chicks fed autoclaved KFSB or CSB increased and pancreas weight (percentage of body weight) decreased as autoclaving time increased. Slightly less autoclaving time was consistently required to achieve maximum chick performance for KFSB compared with CSB. Less autoclaving time was also required to obtain maximum digestibility of amino acids in KFSB compared with CSB. Urease activity of the soybeans decreased as autoclaving time increased, whereas protein solubility in .2% KOH for CSB did not change consistently in response to heating time. The results of the present study indicate that raw KFSB must be heated to obtain maximum protein quality for chicks and that protein solubility in KOH is not a sensitive indicator of underprocessing of CSB soybeans. (Key words: soybeans, Kunitz trypsin inhibitor, protein solubility, urease, chickens)
HEATING EFFECTS ON SOYBEANS VARYING IN TRYPSIN INHIBITOR
MATERIALS AND METHODS
Raw, full-fat CSB2 (Williams-82), KFSB,2 and commercial solvent-extracted dehulled SBM2 were ground to a similar particle size. Both CSB and KFSB contained 12% moisture. The CSB and KFSB were placed in a thin layer (1.25 cm in depth) on metal trays and covered with aluminum foil. The two types of soybeans were then cooked simultaneously within experiment from 0 to 21 min in 3-min increments in a laboratory autoclave at 121 C and 124 kPa. Length of autoclaving varied among experiments. The autoclaving time periods began when the temperature reached 121 C, which occurred within 30 s after the autoclave was turned on. The soybeans in Experiments 1, 2, 3, and 5 were cooked in the same
2 The CSB and KFSB were obtained from Illinois Foundation Seeds, Inc., Champaign, IL 61820, and SBM was obtained from Central Soya, Inc., Gibson City, IL 60936.
autoclave and those in Experiment 4 were cooked in a different laboratory autoclave. The SBM used in all experiments came from the same lot. Analytical Methods for Urease Activity, Protein Solubility, and Trypsin Inhibitor
Urease activity of the autoclaved soybeans was determined by the pH change method described by the American Oil Chemists Society (1980). Dry matter and Kjeldahl nitrogen analyses were conducted by methods outlined by the Association of Official Analytical Chemists (1984). Protein solubility of CSB was determined by the method described in Araba and Dale (1990a) and Parsons et al. (1991). The soybeans were ground to a mean particle size of approximately 200 urn prior to analysis. Trypsin inhibitor analyses were conducted on CSB, KFSB, and SBM using the enzyme assay described by Friedman et al. (1991) and expressed as trypsin inhibitor units per gram of soybeans or SBM. A trypsin inhibitor unit is the reduction in activity of trypsin by one trypsin unit, and a trypsin unit is defined as the amount of trypsin that catalyzes the hydrolysis of 1 umol of substrate/min (Friedman et al, 1991). Chick Assays
Five chick assays were conducted to evaluate the effects of autoclaving on in vivo protein quality of the soybeans. All experiments used 8- or 9-day-old male chicks resulting from the cross of New Hampshire males and Columbian Plymouth Rock females. The chicks were housed in thermostatically controlled starter batteries with raised wire floors and allowed ad libitum access to water and feed. Uniform light was provided 24 h daily. The chicks were fed a 24% CP corn and SBM pretest diet during the first 7 days posthatching. Chicks were then fasted overnight, weighed, wing-banded, and allotted to dietary treatments by the procedure of Sasse and Baker (1974). Dietary treatments consisted of feeding the autoclaved CSB or KFSB as the sole source of dietary protein in 23% CP dex-
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
because KFSB contain substantially less total trypsin inhibitor than CSB (Friedman et al, 1991; Han et al, 1991), the amount of heat or cooking time required to produce optimum protein quality may be less for KFSB than for CSB. Protein solubility in .2% KOH has been shown recently to be a good indicator of overprocessing of SBM (Araba and Dale, 1990a; Parsons et al, 1991). Araba and Dale (1990b) reported that this protein solubility assay may also be useful in detecting underprocessing of SBM but its sensitivity and accuracy as an index of underprocessing have not been elucidated clearly. Quality-control laboratories would benefit from having one test that would detect both under- and overprocessed SBM. The objectives of the present study were 1) to determine whether steam heating would improve the protein quality of raw KFSB for chickens and whether the amount of heating required to produce optimal protein quality is different for KFSB than CSB, and 2) to evaluate protein solubility in .2% KOH as an in vitro indicator of in vivo protein quality for underprocessed soybeans.
1701
1702
ANDERSON-HAFERMANN ET AL. TABLE 1. Composition of experimental diets
Ingredients
Conventional soybean diet
Kunitz trypsin inhibitor-free soybean diet
Soybean source1 Dextrose Soybean oil Cellulose Dicalcium phosphate Ground limestone Iodized salt DL-methionine Vitamin mix2 CholineCl (60%) Trace minerals mix3
62.20 33.75 ... ... 2.20 1.00 .40 .20 .10 .10 .05
63.00 32.95 ... ... 2.20 1.00 .40 .20 .10 .10 .05
Dehulled soybean meal diet
(%)
% w conventional full-fat soybeans contained 37.0% CP, raw Kunitz trypsin inhibitor-free full-fat soybeans contained 36.5% CP, and dehulled solvent-extracted soybean meal contained 46.3% CP. 2 Vitamin mix provided per kilogram of diet: vitamin A (as vitamin A acetate), 4,400 IU; cholecalciferol, 1,000 IU; vitamin E (as a-tocopheryl acetate), 11 IU; vitamin Bj2, -01 mg; riboflavin, 4.41 mg; d-pantothenic acid, 10.0 mg; niacin, 22.0 mg; menadione sodium bisulfite, 2.33 mg. 3 Trace minerals mix contained provided the following in milligrams per kilogram of diet: Mn, 75.0 from manganese oxide; Fe, 75.0 from iron sulfate; Zn, 75.0 from zinc oxide; Cu, 5.0 from copper sulfate; I, .75 from ethylene diamine dihydroiodide; Se, .10 from sodium selenite.
trose and soybean diets (Table 1) formulated to meet all nutrient requirements of broiler chicks (National Research Council, 1984). A dextrose and SBM diet (23% CP) was also included in most chick assays for comparison. Soybean oil and cellulose were added to the latter diet to approximate the digestible oil and fiber contents of the CSB and KFSB diets. Experiment 1 was an initial chick assay to evaluate the effect of autoclaving on KFSB. Experiments 2 to 5 evaluated and compared the effects of autoclaving on both CSB and KFSB. Because CSB and KFSB were evaluated separately in Experiments 2 and 3, respectively, a treatment consisting of CSB autoclaved for 6 min was included in Experiment 3 to assess the effects of experiment (or time) when comparing the results of Experiments 2 and 3. Each of the diets in the five experiments were fed to triplicate groups of six male chicks from 8 or 9 to 17 or 18 days posthatching, respectively. Body weight and feed intake of each group were measured at the termination of the experiments, and weight gain and feed efficiency (gain: feed ratio) were calculated. At the termination of most experiments, chicks were killed by cervical dislocation, and the pancreas
removed and weighed, with the pancreas weight expressed as a percentage of body weight. Amino Acid Digestibility
Assay
Cecectomized (CEC) Single Comb White Leghorn cockerels, 40 wk of age, were used to determine true digestibility of amino acids in CSB or KFSB autoclaved for 0,9, or 18 min. The CSB and KFSB samples were the same as those fed to chicks in Experiments 2 and 3, respectively. The birds were kept in individual cages with raised wire floors in an environmentally regulated room and provided with 16 h of light daily. Feed and water were consumed ad libitum before the initiation of the experiment. Cecectomy was performed at 20 wk of age according to the procedure of Parsons (1985). Previous work in the authors' laboratory indicated that little or no cecal regrowth occurred within 1 yr following the surgery (Parsons, 1985). The digestibility assay was the same as that described by Sibbald (1979), with some minor modifications. Following a 24-h fast, four CEC cockerels were given 30 g of soybeans via crop intubation. Four additional cockerels were fasted throughout the experimental period to measure en-
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
49.70 34.45 9.10 2.70 2.20 1.00 .40 .20 .10 .10 .05
HEATING EFFECTS ON SOYBEANS VARYING IN TRYPSIN INHIBITOR
1703
TABLE 2. Effect of autoclaving on nutritional quality of Kunitz trypsin inhibitor-free soybeans in Experiment 1 Weight gain1-2
Gain: feed1-2
Pancreas weight2
(min) 0 3 6 9 12 18 SBM4 SEM
(g) 120 119 134 141 152 151 163 4.7
(g=g) .501 .491 .596 .607 .656 .668 .711 .017
(% BW) .655 .669 .577 .558 .513 .420 .337 .018
Urease index3 (units of pH increase) 2.2 2.2 2.2 2.2 1.1 .6 .2 .03
^Values are means of three replicate groups of six male chicks from 8 to 17 days of age; average initial weight was 76.5 g. 2 Linear effect of autoclaving (P < .001). 3 Values are means of duplicate analyses. 4 SBM = commercial dehulled soybean meal.
dogenous excretion of amino acids. A plastic tray was placed under each cage, and excreta were collected for 48 h. Excreta samples were lyophilized, weighed, and ground to pass through a 60-mesh screen. Amino acid concentrations of the soybeans and excreta samples were analyzed by ionexchange chromatography3 (Spackman et al, 1958) following hydrolysis in 6N HCl for 22 h at 110 C in evacuated sealed tubes, with at least two replicates per sample. Methionine and cystine were determined on samples that had been oxidized in performic acid prior to acid hydrolysis according to the procedure of Moore (1963), with the exception that the excess performic acid was removed by lyophilization after dilution with water. Statistical Analysis
weight gain, feed efficiency, pancreas weight as a percentage of body weight, or protein solubility as the dependent variables. Pancreas weights (grams) were also analyzed by analysis of covariance with body weight as described by Brown et al. (1985). The results of the analysis of variance for pancreas weight as a percentage of body weight were almost identical and not significantly different from the results obtained by the analysis of covariance. Therefore, the pancreas data are presented as percentage of body weight in the data tables for simplicity. Statistical significance of differences among true amino acid digestibility values was assessed with the least significance difference test (Steel and Torrie, 1980). RESULTS
The results of Experiment 1, in which chicks were fed KFSB, are shown in Table 2. Increasing autoclaving time from 0 to 12 min resulted in a linear increase (P < .001) in weight gain and feed efficiency, with no further significant improvement being observed for 18 min. Pancreas weight decreased linearly (P < .001) as autoclaving time was increased from 0 to 18 min. Performance of chicks fed KFSB au3 Amino acid analyses performed with a Beckman toclaved for 18 min was not significantly 6300 Analyzer, Beckman Instruments Corp., Palo Alto, different (P > .05) from performance of chicks fed SBM, whereas pancreas weight CA 94302.
Data from chick growth assays were subjected to analysis of variance procedures for completely randomized designs (Steel and Torrie, 1980). Regression analysis (Steel and Torrie, 1980) was used to evaluate the chick performance data, with autoclaving time as the independent variable and
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
Autoclaving time
1704
ANDERSON-HAFERMANN ET AL.
TABLE 3. Effect of autoclaving on nutritional quality of conventional soybeans in Experiment 2 Weight gain1-2
Gain: feed1-2
Pancreas weight2
Protein solubility3
(min) 0 3 6 9 12 15 18 SBM5 SEM
(g) 98 113 120 124 143 150 151 158 3.6
(g=g) .481 .500 .518 .567 .621 .682 .662 .680 .019
(% BW) .880 NM4 .847 NM .562 NM .392 .372 .017
(%) 87 89 91 91 87 85 76 77 .6
Urease index3 (units of pH increase) 2.2 2.2 2.1 1.9 .2 0 .1 .2 .01
Trypsin inhibitor3 (units per gram sample) 6,780 ± 394 NM 5,842 ± 160 3,548 ± 66 2,179 ± 40 1,169 ± 103 NM 268 ± 6
Values are means of three replicate groups of six male chicks from 8 to 17 days of age; average initial weight was 68.7 g. 2 Linear effect of autoclaving (P < .001). 3 Protein solubility and urease index values are means of duplicate analyses and trypsin inhibitor values are 5c ± SD for triplicate analyses. 4 NM = not measured. 5SBM = commercial dehulled soybean meal.
of chicks fed KFSB autoclaved for 18 min was greater (P < .05) than pancreas weight of chicks fed SBM. The urease index of KFSB autoclaved for 0 to 9 min was high but then decreased markedly at 12 and 18 min of autoclaving. The effects of autoclaving on CSB in Experiment 2 are shown in Table 3. As autoclaving time increased from 0 to 15 min, chick performance increased linearly (P < .001). No further improvement in weight gain or feed efficiency was obtained at 18 min. A linear decrease (P < .001) in pancreas weight was observed as autoclaving time increased from 0 to 18 min. Weight gain and feed efficiency of chicks fed CSB autoclaved for 15 or 18 min and pancreas weight of chicks fed CSB autoclaved for 18 min were not significantly different (P > .05) from those of chicks fed SBM. Protein solubility did not change substantially or consistently as autoclaving time increased from 0 to 15 min. The urease index of CSB autoclaved for 0 to 9 min was approximately two pH units and decreased markedly by 12 min of autoclaving. Trypsin inhibitor content of CSB decreased as autoclaving time increased. The trypsin inhibitor content of raw CSB was 25-fold higher than SBM,
and the trypsin inhibitor content of CSB autoclaved for 15 min was 4-fold higher than SBM. The effects of autoclaving KFSB in Experiment 3 (Table 4) were similar to those observed in Experiment 1. Chick growth performance increased linearly (P < .001) as autoclaving time increased to 12 min, with no additional response to 15 or 18 min of autoclaving. Pancreas weight decreased (P < .001) as autoclaving time increased from 0 to 15 min. Growth performance and pancreas weight of chicks fed CSB autoclaved for 6 min were similar to those observed for the same treatment in Experiment 2 and were inferior to those of chicks fed KFSB autoclaved for 6 min. Feed efficiencies, but not weight gains, of chicks fed KFSB autoclaved for 12 to 18 min were significantly lower (P < .05) than feed efficiency of chicks fed SBM. The urease index decreased markedly at 9 and 12 min of autoclaving. Trypsin inhibitor content of KFSB decreased as autoclaving time increased and was much lower (P < .05) than trypsin inhibitor content of CSB autoclaved for the same amount of time in Experiment 2 (Table 3). Trypsin inhibitor content of KFSB autoclaved for 15 min was greater (P < .01) than that of SBM.
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
Autoclaving time
HEATING EFFECTS ON SOYBEANS VARYING IN TRYPSIN INHIBITOR
1705
TABLE 4. Effect of autoclaving on nutritional quality of Kunitz trypsin inhibitor-free soybeans in Experiment 3 Weight gain1-2
Gain: feed1-2
Pancreas weight2
(min) 0 3 6 9 12 15 18 65 SBM« SEM
(g)
(g:g) .559 .576 .632 .637 .691 .717 .685 .509 .751 .012
(% BW) .640 .650 .604 .519 .449 .409 .393 .833 .326 .018
129 131 138 149 158 161 157 110 167 3.4
Urease index3 (units of pH increase) 2.2 2.2 2.1 1.0 .3 .1 .1 2.1 .2 .01
Trypsin inhibitor3 (units per gram sample) 4,276 ± 222 NM* 3,012 ± 72 1,812 ± 72 844 ± 25 469 ± 26 NM 5,842 ± 160 268 ± 6
Walues are means of three replicate groups of six male chicks from 8 to 17 days of age; average initial weight was 69.0 g. 2 Linear effect of autoclaving (P < .001). 3 Urease index values are means of duplicate analyses and trypsin inhibitor values are x ± SD for triplicate analyses. 4 NM = not measured. 5 Conventional soybeans autoclaved for 6 min. Same soybeans as evaluated in Experiment 2 (Table 3). *SBM = commercial dehulled soybean meal.
Chicks fed raw KFSB gained more rapidly and efficiently and had smaller pancreas weights than those fed raw CSB in Experiment 4 (Table 5). Increasing autoclaving time resulted in a significant linear increase (P < .05) in weight gain and feed efficiency for chicks fed either CSB or KFSB; however, weight gain did not increase significantly (P > .05) beyond 12 min of autoclaving. Increasing autoclaving time resulted in a linear decrease (P < .001) in pancreas weight for chicks fed CSB and a quadratic decrease (P < .01) in pancreas weight for chicks fed KFSB. Maximal feed efficiency and minimal pancreas weight were obtained by 18 min of autoclaving for CSB and by 15 min of autoclaving for KFSB. Protein solubility of CSB did not change consistently as autoclaving time increased from 0 to 18 min. There was a 16% decrease in protein solubility for CSB at 21 min compared with 18 min of autoclaving. The urease index of CSB and KFSB autoclaved for 0 to 9 min was high and then decreased at 12 and 15 min of autoclaving. Experiment 5 (Table 6) was conducted to further evaluate and compare the
effects of autoclaving on CSB versus KFSB during the 0 to 12-min period, which was in the linear part of the growth response curve observed in the previous experiments. At 0 min of autoclaving, weight gain and feed efficiency of chicks consuming CSB were inferior (P < .05) to those of chicks consuming KFSB. Growth performance increased linearly as autoclaving time increased for both soybean types. Large differences in weight gain and feed efficiency between CSB and KFSB were still evident at 12 min of autoclaving, and weight gain of chicks consuming CSB autoclaved for 12 min was much below their maximum growth rate. Linear regression analysis of weight gain on autoclaving time indicated that the slope of the regression lines for CSB and KFSB were almost identical (Y = 96.2 + 2.6X, r = .99 for CSB and Y = 117.5 + 2.7X, r = .99 for KFSB). True digestibilities of some amino acids in CSB and KFSB autoclaved for 0, 9, or 18 min are shown in Table 7. Digestibility of amino acids in KFSB autoclaved for 0 or 9 min was substantially higher than digestibility of amino acids in CSB autoclaved for 0 or 9 min, with no difference between
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
Autoclaving time
1706
ANDERSON-HAFERMANN ET AL.
TABLE 5. Effect of autoclaving on nutritional quality of conventional soybeans (CSB) and Kunitz trypsin inhibitor-free soybeans (KFSB) in Experiment 4 Soybean type
CSB
SBM5 SEM
(min) 0 6 12 15 18 21 0 6 12 15 18 21
—
Weight gain1-2
(g) 122 124 152 153 155 156 135 140 150 152 153 155 156 2.9
Gain: feed1-2
(g:g) .505 .542 .615 .632 .652 .667 .576 .606 .626 .654 .657 .670 .685 .011
Pancreas weight2
(% BW) .854 .762 .483 .420 .330 .344 .631 .557 .376 .341 .335 .334 .323 .012
Protein solubility3
(%) 93 86 90 90 90 74 NM* NM NM NM NM NM 74 1.4
Urease index3 (units of pH increase) 2.5 2.4 1.4 .1 0 0 2.5 2.5 1.6 .2 0 0 .1 .05
1 Values are means of three replicate groups of six male chicks from 8 to 17 days of age; average initial weight was 102.3 g. Significant linear effect (P < .05) of autoclaving for weight gain, gain:feed ratio, and pancreas weight for chicks fed CSB and for weight gain and gain:feed ratio for chicks fed KFSB. Quadratic effect (P < .01) of autoclaving for pancreas weight of chicks fed KFSB soybeans. 3 Values are means of duplicate analyses. 4 NM = not measured. 5SBM = commercial dehulled soybean meal.
Zhang et al, 1991), which indicated that the nutritive value of raw KFSB for chickens is greater than that of raw CSB. However, the studies conducted herein further show that heating of KFSB is required to obtain maximum nutritive value for chicks. Friedman et al. (1991) reported similar results for rats. The need for heating of KFSB is probably largely due to its trypsin inhibitor content. Although raw KFSB contained 40 to 50% less total trypsin inhibitor activity than raw CSB, the level of trypsin inhibitor in raw KFSB was approximately 16-fold higher than that in SBM. The KFSB contain levels of Bowman-Birk protease inhibitors that are similar to those in CSB (Friedman et al, 1991). Some of the beneficial effects of heating KFSB were also probably associated with denaturation of the protein (Kakade et al, 1973) and reduction in hemagglutinating acDISCUSSION tivity (Friedman et al, 1991). The duration of autoclaving or heating The results of the present study confirm those of previous reports (Han et al, 1991; needed to maximize chick weight gain
soybean types at 18 min. Also, with the exception of methionine, there were no significant (P > .05) differences between digestibilities of amino acids in KFSB autoclaved for 9 min and CSB autoclaved for 18 min. Maximum digestibility of amino acids in KFSB was obtained by 9 min of autoclaving whereas autoclaving in excess of 9 min was required to obtain maximum digestibility of amino acids in CSB. Autoclaving KFSB for 9 min increased (P < .05) digestibility of methionine and cystine compared with 0 min, with no further improvement from 18 min of autoclaving. Although digestibility of other amino acids in KFSB were increased by approximately five percentage units by autoclaving for 9 min compared with 0 min, these increases were not significant (P > .05).
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
KFSB
Autoclaving time
HEATING EFFECTS ON SOYBEANS VARYING IN TRYPSIN INHIBITOR
1707
TABLE 6. Effect of autoclaving on nutritional quality of conventional soybeans (CSB) and Kunitz trypsin inhibitor-free soybeans (KFSB) in Experiment 5 Soybean type
CSB KFSB
Weight gain1-2
Gain: feed1-2
Protein solubility3
(min) 0 6 12 0 6 12
(g) 97 110 128 117 135 150 3.1
(g:g) .421 .469 .577 .514 .571 .634 .008
(%) 93 92 92 NM* NM NM .9
Urease index3 (units of pH increase) 2.2 2.1 1.1 2.2 2.1 1.1 .02
1 Values are means of three replicate groups of six male chicks from 9 to 18 days of age; average initial weight was 92.1 g. Significant linear effect of autoclaving (P < .001) for both soybean types. 3 Values are means of duplicate analyses. 4 NM = not measured.
and feed efficiency and minimize pancreatic weight was about 15 to 20% less for KFSB than CSB in Experiments 1 to 4. Regression analysis of the data in chick Experiment 5 also predicted that maximum weight gain of chicks fed KFSB would be obtained at less heating time than chicks fed CSB. Less heating time was also required to obtain maximum amino acid digestibility for KFSB compared with CSB. The difference in required heating time between soybean types was probably largely because of the lower level of trypsin inhibitor in KFSB. The results of Experiments 2 and 3 showed that maximum chick performance was obtained when the trypsin inhibitor content of CSB was reduced to 1,169 units/g by 15 min of autoclaving and when the trypsin inhibitor content of KFSB was reduced to 844 units/g by 12 min of autoclaving. Friedman et cd. (1991) reported that soy flour supported optimum rat growth when the trypsin inhibitor activity was reduced to about 1,000 units/g. Although less heating time was consistently required for KFSB versus CSB to maximize chick performance, the magnitude of this difference (15 to 20%) was less than the difference in trypsin inhibitor content (41%) between the soybean types. Thus, it seems that other factors such as denaturation of the soybean protein and
destruction of hemagglutinating factors (discussed earlier) may influence the amount of heating needed to optimize protein quality. Although CSB autoclaved for 15 min (Experiment 2) and KFSB autoclaved for 12 min (Experiment 3) still contained substantially more trypsin inhibitor than did commercial SBM, they both yielded chick growth performance that was similar to that obtained from SBM. Thus, the level of trypsin inhibitor in these CSB and KFSB diets was apparently low enough that it did not depress growth. The second main objective of the current study was to evaluate protein solubility in .2% KOH as an indicator of underprocessed CSB. Although increased autoclaving of CSB for 0 to 12, 15, or 18 min improved chick growth performance greatly among experiments, protein solubility did not change substantially or consistently as autoclaving time increased. Araba and Dale (1990b) reported that the protein solubility assay was of value in detecting underprocessed SBM. However, the results of the current study indicated that protein solubility in .2% KOH is not a sensitive index of underprocessing of CSB. The difference in results between these two studies is probably not due to differences in soybean type evaluated (SBM versus CSB) because Dale and Araba (1990) reported that oil content of soybe-
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
SEM
Autoclaving time
1708
ANDERSON-HAFERMANN ET AL.
TABLE 7. Effect of autoclaving on true digestibility of some amino acids in conventional soybeans and Kunitz trypsin inhibitor-free soybeans in Experiment 6 1 Amino acid
CSB 0
KFSB 0
True digestibility* CSB 9 KFSB 9
64
1992 Poultry Science 71:1700-1709
that the nutritional value of raw Kunitz trypsin inhibitor-free soybeans (KFSB) was A large amount of research has been substantially higher than that of raw conducted to evaluate the detrimental conventional soybeans (CSB) for chicks effects on animal performance caused by and laying hens but was less than that of antinutritional factors in raw soybeans (Friedman et ah, 1991). Kunitz trypsin commercial soybean meal (SBM). The inhibitor is one of the major antinutritional KFSB is a soybean variant that is isogenic factors in soybeans (Rackis, 1965). Han et to the commercial Williams 82 cultivar al. (1991) and Zhang et al. (1991) reported except that it lacks a functional Kunitz trypsin inhibitor allele (Bernard and Hymowitz, 1986). Friedman et al. (1991) recently showed that steam heating of Received for publication February 24, 1992. KFSB increased its protein quality for rats. Accepted for publication June 5, 1992. These results suggest that some heating of J To whom correspondence should be addressed: 322 Mumford Hall, 1301 West Gregory Drive, Urbana, KFSB is probably necessary to obtain maximal quality for poultry. However, IL 61801. INTRODUCTION
1700
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
ABSTRACT Five 10-day chick growth experiments and an amino acid digestibility assay were conducted to assess the effect of steam heating on in vivo protein quality of raw, full-fat Kunitz trypsin inhibitor-free soybeans (KFSB) compared with raw, conventional full-fat soybeans (CSB). Protein solubility in .2% KOH was also evaluated as an in vitro test of in vivo protein quality for underprocessed CSB. The CSB and KFSB were autoclaved for 0,3,6, 9,12,15,18, or 21 min at 121 C and 124 kPa. The soybeans were then fed to 8- or 9-day-old chicks as the sole source of protein in dextrose and soybean diets containing 23% protein. Growth performance of chicks fed raw KFSB was superior to that of chicks fed raw CSB. Growth performance of chicks fed autoclaved KFSB or CSB increased and pancreas weight (percentage of body weight) decreased as autoclaving time increased. Slightly less autoclaving time was consistently required to achieve maximum chick performance for KFSB compared with CSB. Less autoclaving time was also required to obtain maximum digestibility of amino acids in KFSB compared with CSB. Urease activity of the soybeans decreased as autoclaving time increased, whereas protein solubility in .2% KOH for CSB did not change consistently in response to heating time. The results of the present study indicate that raw KFSB must be heated to obtain maximum protein quality for chicks and that protein solubility in KOH is not a sensitive indicator of underprocessing of CSB soybeans. (Key words: soybeans, Kunitz trypsin inhibitor, protein solubility, urease, chickens)
HEATING EFFECTS ON SOYBEANS VARYING IN TRYPSIN INHIBITOR
MATERIALS AND METHODS
Raw, full-fat CSB2 (Williams-82), KFSB,2 and commercial solvent-extracted dehulled SBM2 were ground to a similar particle size. Both CSB and KFSB contained 12% moisture. The CSB and KFSB were placed in a thin layer (1.25 cm in depth) on metal trays and covered with aluminum foil. The two types of soybeans were then cooked simultaneously within experiment from 0 to 21 min in 3-min increments in a laboratory autoclave at 121 C and 124 kPa. Length of autoclaving varied among experiments. The autoclaving time periods began when the temperature reached 121 C, which occurred within 30 s after the autoclave was turned on. The soybeans in Experiments 1, 2, 3, and 5 were cooked in the same
2 The CSB and KFSB were obtained from Illinois Foundation Seeds, Inc., Champaign, IL 61820, and SBM was obtained from Central Soya, Inc., Gibson City, IL 60936.
autoclave and those in Experiment 4 were cooked in a different laboratory autoclave. The SBM used in all experiments came from the same lot. Analytical Methods for Urease Activity, Protein Solubility, and Trypsin Inhibitor
Urease activity of the autoclaved soybeans was determined by the pH change method described by the American Oil Chemists Society (1980). Dry matter and Kjeldahl nitrogen analyses were conducted by methods outlined by the Association of Official Analytical Chemists (1984). Protein solubility of CSB was determined by the method described in Araba and Dale (1990a) and Parsons et al. (1991). The soybeans were ground to a mean particle size of approximately 200 urn prior to analysis. Trypsin inhibitor analyses were conducted on CSB, KFSB, and SBM using the enzyme assay described by Friedman et al. (1991) and expressed as trypsin inhibitor units per gram of soybeans or SBM. A trypsin inhibitor unit is the reduction in activity of trypsin by one trypsin unit, and a trypsin unit is defined as the amount of trypsin that catalyzes the hydrolysis of 1 umol of substrate/min (Friedman et al, 1991). Chick Assays
Five chick assays were conducted to evaluate the effects of autoclaving on in vivo protein quality of the soybeans. All experiments used 8- or 9-day-old male chicks resulting from the cross of New Hampshire males and Columbian Plymouth Rock females. The chicks were housed in thermostatically controlled starter batteries with raised wire floors and allowed ad libitum access to water and feed. Uniform light was provided 24 h daily. The chicks were fed a 24% CP corn and SBM pretest diet during the first 7 days posthatching. Chicks were then fasted overnight, weighed, wing-banded, and allotted to dietary treatments by the procedure of Sasse and Baker (1974). Dietary treatments consisted of feeding the autoclaved CSB or KFSB as the sole source of dietary protein in 23% CP dex-
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
because KFSB contain substantially less total trypsin inhibitor than CSB (Friedman et al, 1991; Han et al, 1991), the amount of heat or cooking time required to produce optimum protein quality may be less for KFSB than for CSB. Protein solubility in .2% KOH has been shown recently to be a good indicator of overprocessing of SBM (Araba and Dale, 1990a; Parsons et al, 1991). Araba and Dale (1990b) reported that this protein solubility assay may also be useful in detecting underprocessing of SBM but its sensitivity and accuracy as an index of underprocessing have not been elucidated clearly. Quality-control laboratories would benefit from having one test that would detect both under- and overprocessed SBM. The objectives of the present study were 1) to determine whether steam heating would improve the protein quality of raw KFSB for chickens and whether the amount of heating required to produce optimal protein quality is different for KFSB than CSB, and 2) to evaluate protein solubility in .2% KOH as an in vitro indicator of in vivo protein quality for underprocessed soybeans.
1701
1702
ANDERSON-HAFERMANN ET AL. TABLE 1. Composition of experimental diets
Ingredients
Conventional soybean diet
Kunitz trypsin inhibitor-free soybean diet
Soybean source1 Dextrose Soybean oil Cellulose Dicalcium phosphate Ground limestone Iodized salt DL-methionine Vitamin mix2 CholineCl (60%) Trace minerals mix3
62.20 33.75 ... ... 2.20 1.00 .40 .20 .10 .10 .05
63.00 32.95 ... ... 2.20 1.00 .40 .20 .10 .10 .05
Dehulled soybean meal diet
(%)
% w conventional full-fat soybeans contained 37.0% CP, raw Kunitz trypsin inhibitor-free full-fat soybeans contained 36.5% CP, and dehulled solvent-extracted soybean meal contained 46.3% CP. 2 Vitamin mix provided per kilogram of diet: vitamin A (as vitamin A acetate), 4,400 IU; cholecalciferol, 1,000 IU; vitamin E (as a-tocopheryl acetate), 11 IU; vitamin Bj2, -01 mg; riboflavin, 4.41 mg; d-pantothenic acid, 10.0 mg; niacin, 22.0 mg; menadione sodium bisulfite, 2.33 mg. 3 Trace minerals mix contained provided the following in milligrams per kilogram of diet: Mn, 75.0 from manganese oxide; Fe, 75.0 from iron sulfate; Zn, 75.0 from zinc oxide; Cu, 5.0 from copper sulfate; I, .75 from ethylene diamine dihydroiodide; Se, .10 from sodium selenite.
trose and soybean diets (Table 1) formulated to meet all nutrient requirements of broiler chicks (National Research Council, 1984). A dextrose and SBM diet (23% CP) was also included in most chick assays for comparison. Soybean oil and cellulose were added to the latter diet to approximate the digestible oil and fiber contents of the CSB and KFSB diets. Experiment 1 was an initial chick assay to evaluate the effect of autoclaving on KFSB. Experiments 2 to 5 evaluated and compared the effects of autoclaving on both CSB and KFSB. Because CSB and KFSB were evaluated separately in Experiments 2 and 3, respectively, a treatment consisting of CSB autoclaved for 6 min was included in Experiment 3 to assess the effects of experiment (or time) when comparing the results of Experiments 2 and 3. Each of the diets in the five experiments were fed to triplicate groups of six male chicks from 8 or 9 to 17 or 18 days posthatching, respectively. Body weight and feed intake of each group were measured at the termination of the experiments, and weight gain and feed efficiency (gain: feed ratio) were calculated. At the termination of most experiments, chicks were killed by cervical dislocation, and the pancreas
removed and weighed, with the pancreas weight expressed as a percentage of body weight. Amino Acid Digestibility
Assay
Cecectomized (CEC) Single Comb White Leghorn cockerels, 40 wk of age, were used to determine true digestibility of amino acids in CSB or KFSB autoclaved for 0,9, or 18 min. The CSB and KFSB samples were the same as those fed to chicks in Experiments 2 and 3, respectively. The birds were kept in individual cages with raised wire floors in an environmentally regulated room and provided with 16 h of light daily. Feed and water were consumed ad libitum before the initiation of the experiment. Cecectomy was performed at 20 wk of age according to the procedure of Parsons (1985). Previous work in the authors' laboratory indicated that little or no cecal regrowth occurred within 1 yr following the surgery (Parsons, 1985). The digestibility assay was the same as that described by Sibbald (1979), with some minor modifications. Following a 24-h fast, four CEC cockerels were given 30 g of soybeans via crop intubation. Four additional cockerels were fasted throughout the experimental period to measure en-
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
49.70 34.45 9.10 2.70 2.20 1.00 .40 .20 .10 .10 .05
HEATING EFFECTS ON SOYBEANS VARYING IN TRYPSIN INHIBITOR
1703
TABLE 2. Effect of autoclaving on nutritional quality of Kunitz trypsin inhibitor-free soybeans in Experiment 1 Weight gain1-2
Gain: feed1-2
Pancreas weight2
(min) 0 3 6 9 12 18 SBM4 SEM
(g) 120 119 134 141 152 151 163 4.7
(g=g) .501 .491 .596 .607 .656 .668 .711 .017
(% BW) .655 .669 .577 .558 .513 .420 .337 .018
Urease index3 (units of pH increase) 2.2 2.2 2.2 2.2 1.1 .6 .2 .03
^Values are means of three replicate groups of six male chicks from 8 to 17 days of age; average initial weight was 76.5 g. 2 Linear effect of autoclaving (P < .001). 3 Values are means of duplicate analyses. 4 SBM = commercial dehulled soybean meal.
dogenous excretion of amino acids. A plastic tray was placed under each cage, and excreta were collected for 48 h. Excreta samples were lyophilized, weighed, and ground to pass through a 60-mesh screen. Amino acid concentrations of the soybeans and excreta samples were analyzed by ionexchange chromatography3 (Spackman et al, 1958) following hydrolysis in 6N HCl for 22 h at 110 C in evacuated sealed tubes, with at least two replicates per sample. Methionine and cystine were determined on samples that had been oxidized in performic acid prior to acid hydrolysis according to the procedure of Moore (1963), with the exception that the excess performic acid was removed by lyophilization after dilution with water. Statistical Analysis
weight gain, feed efficiency, pancreas weight as a percentage of body weight, or protein solubility as the dependent variables. Pancreas weights (grams) were also analyzed by analysis of covariance with body weight as described by Brown et al. (1985). The results of the analysis of variance for pancreas weight as a percentage of body weight were almost identical and not significantly different from the results obtained by the analysis of covariance. Therefore, the pancreas data are presented as percentage of body weight in the data tables for simplicity. Statistical significance of differences among true amino acid digestibility values was assessed with the least significance difference test (Steel and Torrie, 1980). RESULTS
The results of Experiment 1, in which chicks were fed KFSB, are shown in Table 2. Increasing autoclaving time from 0 to 12 min resulted in a linear increase (P < .001) in weight gain and feed efficiency, with no further significant improvement being observed for 18 min. Pancreas weight decreased linearly (P < .001) as autoclaving time was increased from 0 to 18 min. Performance of chicks fed KFSB au3 Amino acid analyses performed with a Beckman toclaved for 18 min was not significantly 6300 Analyzer, Beckman Instruments Corp., Palo Alto, different (P > .05) from performance of chicks fed SBM, whereas pancreas weight CA 94302.
Data from chick growth assays were subjected to analysis of variance procedures for completely randomized designs (Steel and Torrie, 1980). Regression analysis (Steel and Torrie, 1980) was used to evaluate the chick performance data, with autoclaving time as the independent variable and
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
Autoclaving time
1704
ANDERSON-HAFERMANN ET AL.
TABLE 3. Effect of autoclaving on nutritional quality of conventional soybeans in Experiment 2 Weight gain1-2
Gain: feed1-2
Pancreas weight2
Protein solubility3
(min) 0 3 6 9 12 15 18 SBM5 SEM
(g) 98 113 120 124 143 150 151 158 3.6
(g=g) .481 .500 .518 .567 .621 .682 .662 .680 .019
(% BW) .880 NM4 .847 NM .562 NM .392 .372 .017
(%) 87 89 91 91 87 85 76 77 .6
Urease index3 (units of pH increase) 2.2 2.2 2.1 1.9 .2 0 .1 .2 .01
Trypsin inhibitor3 (units per gram sample) 6,780 ± 394 NM 5,842 ± 160 3,548 ± 66 2,179 ± 40 1,169 ± 103 NM 268 ± 6
Values are means of three replicate groups of six male chicks from 8 to 17 days of age; average initial weight was 68.7 g. 2 Linear effect of autoclaving (P < .001). 3 Protein solubility and urease index values are means of duplicate analyses and trypsin inhibitor values are 5c ± SD for triplicate analyses. 4 NM = not measured. 5SBM = commercial dehulled soybean meal.
of chicks fed KFSB autoclaved for 18 min was greater (P < .05) than pancreas weight of chicks fed SBM. The urease index of KFSB autoclaved for 0 to 9 min was high but then decreased markedly at 12 and 18 min of autoclaving. The effects of autoclaving on CSB in Experiment 2 are shown in Table 3. As autoclaving time increased from 0 to 15 min, chick performance increased linearly (P < .001). No further improvement in weight gain or feed efficiency was obtained at 18 min. A linear decrease (P < .001) in pancreas weight was observed as autoclaving time increased from 0 to 18 min. Weight gain and feed efficiency of chicks fed CSB autoclaved for 15 or 18 min and pancreas weight of chicks fed CSB autoclaved for 18 min were not significantly different (P > .05) from those of chicks fed SBM. Protein solubility did not change substantially or consistently as autoclaving time increased from 0 to 15 min. The urease index of CSB autoclaved for 0 to 9 min was approximately two pH units and decreased markedly by 12 min of autoclaving. Trypsin inhibitor content of CSB decreased as autoclaving time increased. The trypsin inhibitor content of raw CSB was 25-fold higher than SBM,
and the trypsin inhibitor content of CSB autoclaved for 15 min was 4-fold higher than SBM. The effects of autoclaving KFSB in Experiment 3 (Table 4) were similar to those observed in Experiment 1. Chick growth performance increased linearly (P < .001) as autoclaving time increased to 12 min, with no additional response to 15 or 18 min of autoclaving. Pancreas weight decreased (P < .001) as autoclaving time increased from 0 to 15 min. Growth performance and pancreas weight of chicks fed CSB autoclaved for 6 min were similar to those observed for the same treatment in Experiment 2 and were inferior to those of chicks fed KFSB autoclaved for 6 min. Feed efficiencies, but not weight gains, of chicks fed KFSB autoclaved for 12 to 18 min were significantly lower (P < .05) than feed efficiency of chicks fed SBM. The urease index decreased markedly at 9 and 12 min of autoclaving. Trypsin inhibitor content of KFSB decreased as autoclaving time increased and was much lower (P < .05) than trypsin inhibitor content of CSB autoclaved for the same amount of time in Experiment 2 (Table 3). Trypsin inhibitor content of KFSB autoclaved for 15 min was greater (P < .01) than that of SBM.
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
Autoclaving time
HEATING EFFECTS ON SOYBEANS VARYING IN TRYPSIN INHIBITOR
1705
TABLE 4. Effect of autoclaving on nutritional quality of Kunitz trypsin inhibitor-free soybeans in Experiment 3 Weight gain1-2
Gain: feed1-2
Pancreas weight2
(min) 0 3 6 9 12 15 18 65 SBM« SEM
(g)
(g:g) .559 .576 .632 .637 .691 .717 .685 .509 .751 .012
(% BW) .640 .650 .604 .519 .449 .409 .393 .833 .326 .018
129 131 138 149 158 161 157 110 167 3.4
Urease index3 (units of pH increase) 2.2 2.2 2.1 1.0 .3 .1 .1 2.1 .2 .01
Trypsin inhibitor3 (units per gram sample) 4,276 ± 222 NM* 3,012 ± 72 1,812 ± 72 844 ± 25 469 ± 26 NM 5,842 ± 160 268 ± 6
Walues are means of three replicate groups of six male chicks from 8 to 17 days of age; average initial weight was 69.0 g. 2 Linear effect of autoclaving (P < .001). 3 Urease index values are means of duplicate analyses and trypsin inhibitor values are x ± SD for triplicate analyses. 4 NM = not measured. 5 Conventional soybeans autoclaved for 6 min. Same soybeans as evaluated in Experiment 2 (Table 3). *SBM = commercial dehulled soybean meal.
Chicks fed raw KFSB gained more rapidly and efficiently and had smaller pancreas weights than those fed raw CSB in Experiment 4 (Table 5). Increasing autoclaving time resulted in a significant linear increase (P < .05) in weight gain and feed efficiency for chicks fed either CSB or KFSB; however, weight gain did not increase significantly (P > .05) beyond 12 min of autoclaving. Increasing autoclaving time resulted in a linear decrease (P < .001) in pancreas weight for chicks fed CSB and a quadratic decrease (P < .01) in pancreas weight for chicks fed KFSB. Maximal feed efficiency and minimal pancreas weight were obtained by 18 min of autoclaving for CSB and by 15 min of autoclaving for KFSB. Protein solubility of CSB did not change consistently as autoclaving time increased from 0 to 18 min. There was a 16% decrease in protein solubility for CSB at 21 min compared with 18 min of autoclaving. The urease index of CSB and KFSB autoclaved for 0 to 9 min was high and then decreased at 12 and 15 min of autoclaving. Experiment 5 (Table 6) was conducted to further evaluate and compare the
effects of autoclaving on CSB versus KFSB during the 0 to 12-min period, which was in the linear part of the growth response curve observed in the previous experiments. At 0 min of autoclaving, weight gain and feed efficiency of chicks consuming CSB were inferior (P < .05) to those of chicks consuming KFSB. Growth performance increased linearly as autoclaving time increased for both soybean types. Large differences in weight gain and feed efficiency between CSB and KFSB were still evident at 12 min of autoclaving, and weight gain of chicks consuming CSB autoclaved for 12 min was much below their maximum growth rate. Linear regression analysis of weight gain on autoclaving time indicated that the slope of the regression lines for CSB and KFSB were almost identical (Y = 96.2 + 2.6X, r = .99 for CSB and Y = 117.5 + 2.7X, r = .99 for KFSB). True digestibilities of some amino acids in CSB and KFSB autoclaved for 0, 9, or 18 min are shown in Table 7. Digestibility of amino acids in KFSB autoclaved for 0 or 9 min was substantially higher than digestibility of amino acids in CSB autoclaved for 0 or 9 min, with no difference between
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
Autoclaving time
1706
ANDERSON-HAFERMANN ET AL.
TABLE 5. Effect of autoclaving on nutritional quality of conventional soybeans (CSB) and Kunitz trypsin inhibitor-free soybeans (KFSB) in Experiment 4 Soybean type
CSB
SBM5 SEM
(min) 0 6 12 15 18 21 0 6 12 15 18 21
—
Weight gain1-2
(g) 122 124 152 153 155 156 135 140 150 152 153 155 156 2.9
Gain: feed1-2
(g:g) .505 .542 .615 .632 .652 .667 .576 .606 .626 .654 .657 .670 .685 .011
Pancreas weight2
(% BW) .854 .762 .483 .420 .330 .344 .631 .557 .376 .341 .335 .334 .323 .012
Protein solubility3
(%) 93 86 90 90 90 74 NM* NM NM NM NM NM 74 1.4
Urease index3 (units of pH increase) 2.5 2.4 1.4 .1 0 0 2.5 2.5 1.6 .2 0 0 .1 .05
1 Values are means of three replicate groups of six male chicks from 8 to 17 days of age; average initial weight was 102.3 g. Significant linear effect (P < .05) of autoclaving for weight gain, gain:feed ratio, and pancreas weight for chicks fed CSB and for weight gain and gain:feed ratio for chicks fed KFSB. Quadratic effect (P < .01) of autoclaving for pancreas weight of chicks fed KFSB soybeans. 3 Values are means of duplicate analyses. 4 NM = not measured. 5SBM = commercial dehulled soybean meal.
Zhang et al, 1991), which indicated that the nutritive value of raw KFSB for chickens is greater than that of raw CSB. However, the studies conducted herein further show that heating of KFSB is required to obtain maximum nutritive value for chicks. Friedman et al. (1991) reported similar results for rats. The need for heating of KFSB is probably largely due to its trypsin inhibitor content. Although raw KFSB contained 40 to 50% less total trypsin inhibitor activity than raw CSB, the level of trypsin inhibitor in raw KFSB was approximately 16-fold higher than that in SBM. The KFSB contain levels of Bowman-Birk protease inhibitors that are similar to those in CSB (Friedman et al, 1991). Some of the beneficial effects of heating KFSB were also probably associated with denaturation of the protein (Kakade et al, 1973) and reduction in hemagglutinating acDISCUSSION tivity (Friedman et al, 1991). The duration of autoclaving or heating The results of the present study confirm those of previous reports (Han et al, 1991; needed to maximize chick weight gain
soybean types at 18 min. Also, with the exception of methionine, there were no significant (P > .05) differences between digestibilities of amino acids in KFSB autoclaved for 9 min and CSB autoclaved for 18 min. Maximum digestibility of amino acids in KFSB was obtained by 9 min of autoclaving whereas autoclaving in excess of 9 min was required to obtain maximum digestibility of amino acids in CSB. Autoclaving KFSB for 9 min increased (P < .05) digestibility of methionine and cystine compared with 0 min, with no further improvement from 18 min of autoclaving. Although digestibility of other amino acids in KFSB were increased by approximately five percentage units by autoclaving for 9 min compared with 0 min, these increases were not significant (P > .05).
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
KFSB
Autoclaving time
HEATING EFFECTS ON SOYBEANS VARYING IN TRYPSIN INHIBITOR
1707
TABLE 6. Effect of autoclaving on nutritional quality of conventional soybeans (CSB) and Kunitz trypsin inhibitor-free soybeans (KFSB) in Experiment 5 Soybean type
CSB KFSB
Weight gain1-2
Gain: feed1-2
Protein solubility3
(min) 0 6 12 0 6 12
(g) 97 110 128 117 135 150 3.1
(g:g) .421 .469 .577 .514 .571 .634 .008
(%) 93 92 92 NM* NM NM .9
Urease index3 (units of pH increase) 2.2 2.1 1.1 2.2 2.1 1.1 .02
1 Values are means of three replicate groups of six male chicks from 9 to 18 days of age; average initial weight was 92.1 g. Significant linear effect of autoclaving (P < .001) for both soybean types. 3 Values are means of duplicate analyses. 4 NM = not measured.
and feed efficiency and minimize pancreatic weight was about 15 to 20% less for KFSB than CSB in Experiments 1 to 4. Regression analysis of the data in chick Experiment 5 also predicted that maximum weight gain of chicks fed KFSB would be obtained at less heating time than chicks fed CSB. Less heating time was also required to obtain maximum amino acid digestibility for KFSB compared with CSB. The difference in required heating time between soybean types was probably largely because of the lower level of trypsin inhibitor in KFSB. The results of Experiments 2 and 3 showed that maximum chick performance was obtained when the trypsin inhibitor content of CSB was reduced to 1,169 units/g by 15 min of autoclaving and when the trypsin inhibitor content of KFSB was reduced to 844 units/g by 12 min of autoclaving. Friedman et cd. (1991) reported that soy flour supported optimum rat growth when the trypsin inhibitor activity was reduced to about 1,000 units/g. Although less heating time was consistently required for KFSB versus CSB to maximize chick performance, the magnitude of this difference (15 to 20%) was less than the difference in trypsin inhibitor content (41%) between the soybean types. Thus, it seems that other factors such as denaturation of the soybean protein and
destruction of hemagglutinating factors (discussed earlier) may influence the amount of heating needed to optimize protein quality. Although CSB autoclaved for 15 min (Experiment 2) and KFSB autoclaved for 12 min (Experiment 3) still contained substantially more trypsin inhibitor than did commercial SBM, they both yielded chick growth performance that was similar to that obtained from SBM. Thus, the level of trypsin inhibitor in these CSB and KFSB diets was apparently low enough that it did not depress growth. The second main objective of the current study was to evaluate protein solubility in .2% KOH as an indicator of underprocessed CSB. Although increased autoclaving of CSB for 0 to 12, 15, or 18 min improved chick growth performance greatly among experiments, protein solubility did not change substantially or consistently as autoclaving time increased. Araba and Dale (1990b) reported that the protein solubility assay was of value in detecting underprocessed SBM. However, the results of the current study indicated that protein solubility in .2% KOH is not a sensitive index of underprocessing of CSB. The difference in results between these two studies is probably not due to differences in soybean type evaluated (SBM versus CSB) because Dale and Araba (1990) reported that oil content of soybe-
Downloaded from http://ps.oxfordjournals.org/ at Michigan State University on January 29, 2015
SEM
Autoclaving time
1708
ANDERSON-HAFERMANN ET AL.
TABLE 7. Effect of autoclaving on true digestibility of some amino acids in conventional soybeans and Kunitz trypsin inhibitor-free soybeans in Experiment 6 1 Amino acid
CSB 0
KFSB 0
True digestibility* CSB 9 KFSB 9
64
