J. Anim Sci.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by van Kempen, T. A. T. G.
Right arrow Articles by Sewalt, V. J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van Kempen, T. A. T. G.
Right arrow Articles by Sewalt, V. J. H.
J. Anim. Sci. 2006. 84:1387-1395
© 2006 American Society of Animal Science

ANIMAL NUTRITION

Selecting soybean meal characteristics preferred for swine nutrition1

T. A. T. G. van Kempen2, E. van Heugten, A. J. Moeser3, N. S. Muley and V. J. H. Sewalt4

Department of Animal Science and Interdepartmental Nutrition Program, North Carolina State University, Raleigh 27695


   Abstract
Top
 Abstract
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 IMPLICATIONS
 LITERATURE CITED
 
As environmental constraints become more important issues for the animal industry, selecting feed ingredients that yield good animal performance but also minimize environmental impact of animal production becomes critical. The objective of this research was to identify which compositional features would be desirable for soybean meal to maximize nutritional value and minimize animal waste. Eight soybean samples were selected from a database of 72, such that maximal variability for CP, NDF, and ADF content was obtained. Samples were subsequently processed into meal using standardized procedures. In Experiment 1, 8 cannulated pigs were used to determine ileal digestibility following a Latin square design. In Experiment 2, 5 of the samples were used in complete feeds and 10 pigs were used in a crossover Latin square design to determine the total tract digestibility, odorants in fresh and 5-d-old manure, and ammonia emission from manure. Differences up to 6% in ileal DM digestibility and 8% in ileal CP digestibility were observed. This difference was reduced to 1.1% for total tract DM digestibility and 4% for total tract CP digestibility. Differences in odorant concentration were 3-fold and for in vitro ammonia emission were 42%. The only compositional variable with a significant effect on digestibility was stachyose, which negatively affected ileal digestibility of DM (r = –0.80, P = 0.02) and energy (r = –0.73, P = 0.04). None of the compositional variables measured affected ileal CP digestibility. Ileal CP digestibility, however, was correlated with estimated CP fermentation in the large intestine (r = –0.86, P = 0.06) and with in vitro ammonia emission after 48h (r = –0.81, P = 0.09). In conclusion, nutritionally relevant variability exists in soy varieties. Low stachyose content is important for maximizing ileal energy digestibility of soybean meal. Although no compositional variable was identified that explained differences in ileal CP digestibility, maximizing ileal CP digestibility is of interest for maximizing the nutritional value of soybean meal and possibly for reducing ammonia and odor emissions.

Key Words: ammonia • digestibility • odor • soybean meal • swine


   INTRODUCTION
Top
 Abstract
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 IMPLICATIONS
 LITERATURE CITED
 
Soybean meal is the predominant protein source in swine diets in many countries around the world (Cromwell, 2000Go). As such, it has the potential to play a major role in odor, ammonia, and waste production, issues that are of importance for sustainability of the swine industry. A recent study by van Kempen et al. (2002)Go showed that digestibility of soybean meal was rather homogeneous across the United States and in The Netherlands. This lack of variability was attributed to uniform processing conditions and lack of genetic variability in the current soybean varieties.

Although lack of variation is convenient from a quality control standpoint, it hinders efforts to improve efficiency through selection of different varieties of soybeans or through improved processing methods (improved from an animal nutrition standpoint). The objective of this research was to determine if nutritionally relevant variation existed among experimental and commercial varieties of soybean meal and what compositional variables were responsible for this variation.


   MATERIALS AND METHODS
Top
 Abstract
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 IMPLICATIONS
 LITERATURE CITED
 
Soybean Meal Samples
Seventy-two samples of soybeans were assayed for CP, NDF, ADF (Table 1Go), and in vitro DM digestibility (proprietary data). Based on these variables, 8 samples were selected using a principal component analysis (The Unscrambler, Camo, Trondheim, Norway) that represented the variation in this database (by selecting extremes on the principal components). The selected samples included 5 genetically modified strains and 3 conventional samples.


View this table:
[in this window]
[in a new window]
  		Table 1. Composition (%, DM basis) of the Pioneer database based on 72 samples of soybeans
 
These soybeans (500 kg per sample) were subsequently processed into soybean meal at the Texas A& M Oilseed Processing Laboratory. Care was taken to minimize variation due to differences in processing conditions. Processing conditions, in brief, were as follows: Whole soybeans were cracked using Ferrel Ross cracking rolls and aspirated to remove hulls and larger contaminants. Cracked material was passed over a Smico Vibroset Screener to remove whole beans, and then conveyed into a French stack cooker and conditioned for 20 min at 66 ± 6°C. Conditioned cracks were processed using Bauer flaking rolls to yield flakes with a thickness of approximately 0.3 mm. Flaked material was subsequently extracted using a Crown Model 2 countercurrent extractor (capacity of 0.21 m2). Commercial hexane was used at a rate of 1:1 (wt:wt) at a temperature of 43 to 60 ± 6°C, with a retention time of 50 min. Extracted material was fed directly into a Crown desolventizer toaster with its decks set for a meal depth of 7 to 12 cm and a product discharge temperature of 82 to 116 ± 6°C.

The resulting soybean meal samples were analyzed by Pioneer Hybrid (Des Moines, IA) using the following procedures: moisture: AOAC 930.15; crude fat: AOAC 920.39; crude fiber: AOAC 962.09; crude protein (combustion): AOAC 990.03; ash: AOAC 942.05; urease activity: AOCS Ba 9–58; trypsin inhibitor: AOCS Ba 12–75; and protein dispersibility index: AOCS Ba 10–65 (AOAC, 1995Go; AOCS, 1998Go). Oligosaccharides were determined using anion exchange chromataography and phytate was determined by high performance liquid chromatography, using methods developed and validated by Pioneer Hybrid.

Ileal Digestibility Trial
A Latin square design with 8 pigs and 8 diets was used for the ileal digestibility experiment. Gilts were 35 ± 2 kg and were fitted with simple T-cannulas 2 wk before the beginning of the experiment. Pigs were housed in smooth-walled metabolism pens in a controlled environment. Water was provided on an ad libitum basis through a bite nipple. Animal care and procedures were approved by the North Carolina State University Institutional Animal Care and Use Committee.

Experimental diets (Table 2Go) were formulated such that each soybean meal sample was the sole source of protein (16% CP) for each of the diets. Actual soybean meal inclusion rates varied from 31 to 34%. Chromic oxide was added at 0.2% as a marker for the indigestible fraction. These diets were fed twice daily at 12-h intervals at a rate of 45 g/kg of BW0.75 per feeding.


View this table:
[in this window]
[in a new window]
  		Table 2. Composition of the diets used in the ileal digestibility trial1
 
Within each period, a 5-d adaptation period was used for each diet, followed by a 2-d collection of ileal chyme (collected 12 h/d, beginning at the morning feeding and ending at the evening feeding). Chyme was collected into bottles that were screwed onto the cannulas and was frozen within 1 h of collection.

Feed and freeze-dried ileal digesta were analyzed for Cr, AA, and DM at the Experiment Station Chemical Laboratories, University of Missouri—Columbia, according to AOAC (1995)Go procedures, and for energy at North Carolina State University, using an IKA model C5000 bomb calorimeter (IKA, Wilmington, NC). Using these data, apparent ileal digestibility was calculated as described previously (van Kempen et al., 2002Go).

Total Tract Digestibility Trial
Of the 8 soybean meals assayed in the ileal digestibility trial, 5 were selected for inclusion in a total tract digestibility trial. Samples selected were those that were the most variable with respect to composition and ileal digestibility.

Ten crossbred barrows with an initial average BW of 25 ± 2 kg were used. Diets were formulated to have identical apparent ileal digestible lysine to ileal digestible energy ratios (Table 3Go). For soybean meal, digestibility data were obtained from the experiment just described. For corn, digestibility data were from Moeser et al. (2002)Go, whose experiment was conducted simultaneously with the experiment just described and followed the same experimental protocol. Vitamins and minerals were supplied to meet or exceed requirements for 20- to 40-kg pigs (NRC, 1998Go). The 5 diets were assigned to 10 barrows according to a crossover 5 x 5 Latin square design. Pigs were housed in metabolism cages (0.6 x 1.5 m) and given ad libitum access to water. Animal care and use procedures were approved by the North Carolina State University Institutional Animal Care and Use Committee.


View this table:
[in this window]
[in a new window]
  		Table 3. Composition (%, as-fed basis) of the diets used in the total tract digestibility trial1
 
Pigs were fed experimental diets twice daily (0800 and 1500) in mash form at a feeding level equal to 45 g/kg of BW0.75 per meal. Each period (7 d) consisted of a 5-d dietary adaptation period followed by a 2-d quantitative collection of feces and urine.

During the collection periods, feces were collected onto wire screens fixed underneath the metabolism cages. Three times daily, collected feces were weighed and immediately partitioned into 2 equal portions; 1 portion was transferred to 2-L buckets (per pig and collection period) and refrigerated. The remaining portion was pooled by pig and collection period and frozen at –20°C until further chemical analysis was conducted.

Urine was voided on sloped stainless steel trays fixed under the metabolism cages and collected into plastic containers packed in ice to minimize any gaseous loss of N. Ice was replaced twice daily. The quantity of urine was recorded, and one-half of the daily urinary excretion was stored in a 2-L plastic container and refrigerated similar to the fecal collection (per pig and collection period); the remaining sample was pooled by pig and collection period and stored at –20°C for further chemical analyses. At the end of each collection period, refrigerated feces and urine were mixed together and homogenized within respective animal and treatment. A portion of this manure was used for determination of ammonia emission, and the remaining manure was used for analysis of odorants.

Odorant determinations were carried out on manure samples frozen immediately after mixing of urine and feces to prevent fermentation and on manure samples stored at room temperature for 5 d to allow for fermentation. Odorants were determined at Iowa State University using a gas chromatograph-mass spectrometer (Gralapp et al., 2001Go, 2002Go).

Ammonia emission of the manure samples was determined by placing 500 mL of the manure mixture in a rectangular (28 x 9.5 x 6 cm; length x width x height) container (Super Oval 1, Tupperware Co., Orlando, FL). Air was drawn through a flow meter (Cole Palmer, Vernon Hills, IL) at a rate of 1.2 L/min, through the container with manure, and then through a gas dispersion tube (Fisher, Pittsburgh, PA) placed in a 500-mL Erlenmeyer flask containing 400 mL of dilute sulfuric acid (1N) to trap the ammonia released from the manure. This sulfuric acid solution was sampled (1.5 mL) at 12, 24, 36, and 48 h and analyzed for ammonia using the procedure of Willis et al. (1996)Go.

Oven-dried (60°C) feed and fecal samples were analyzed in duplicate for GE content using an adiabatic bomb calorimeter (model C5000, IKA, Wilmington, NC). Nitrogen content of soybean meal samples and feces was assayed by the Kjeldahl method and CP was calculated as Kjeldahl–N x 6.25 (AOAC, 1995Go).

Fermentation rates (disappearance from the large intestines) of CP and DM were calculated based on actual ileal digestibilities of CP and DM for the soybean meal samples and on corn and total tract digestibilities for the complete feeds. Data for corn were obtained in an experiment run simultaneously using identical procedures (Moeser et al., 2002Go; 16.9% of corn CP was indigestible at the ileal level, and 21.6% of corn DM was indigestible at the ileal level). As these values were not directly measured in this experiment, fermentation values were only used to evaluate correlations.

Statistical Analysis
Data were analyzed using SPSS 11.0 (SPSS Inc., Chicago, IL). For digestibility data, the model included animal, period, and diet, and the results are listed based on unconditional analyses. Treatment comparisons are based on main effect comparisons of marginal means. Ammonia emission data were analyzed using a mixed model with time as a repeated measure and feces content in the manure as a covariate. The model included animal, period, and diet as discrete variables. Odor data were analyzed separately for each time point using a model including animal, period, and diet. This method was chosen, as odorants in fresh manure are a result of fermentative activity in the large intestine, whereas odorants in 5-d-old manure are representative of bacterial activity in manure. Although it concerns physically the same sample, the nature of this sample is vastly different and should not be considered a repeated measure. Correlations among variables were determined as Pearson’s correlation coefficients using SPSS and with principal component analysis using The Unscrambler. For these analyses, least squares means per treatment were used.


   RESULTS AND DISCUSSION
Top
 Abstract
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
IMPLICATIONS
 LITERATURE CITED
 
Sample Variation
The composition of commercially grown soybeans is relatively homogeneous (Palmer et al. 1996Go; van Kempen et al., 2002Go). Even in the 72 experimental samples screened for this experiment, the CV for CP, NDF, and ADF was only 3.6, 7.4, and 8.8%, respectively (Table 1Go). As 8 samples were selected for maximal variability in fiber and CP content, it is no surprise that the CV within these 8 samples after processing them into soybean meal was greater than within the entire population (Table 4Go). For example, for ADF, a CV of 26% was observed within the 8 samples of soybean meals as compared with 8.8% in the soybean samples. In contrast, the CV for crude protein was practically identical in the 2 populations (3.6 for the 72 samples and 3.7% for the 8 selected samples), suggesting that processing did not augment variation in CP content and likely even decreased it as variable samples were collected for this study.


View this table:
[in this window]
[in a new window]
  		Table 4. Composition of the soybean meal samples (% as-is, except where noted)
 
Samples G and H, which were mutant lines with lower stachyose and raffinose content (Hitz et al., 2002Go), were responsible not only for a large portion of the variation in fiber variables but also in CP. These samples had a 3 to 5% greater CP content, whereas raffinose was decreased by over 70% and stachyose by 90%. For other samples, the values for the stachyose and raffinose are in agreement with Suna et al. (2003)Go.

After processing, samples A, B, C, and F, which were all Round-up Ready, were similar in composition. Samples D and E were high in fiber, NDF, and ADF. These were commodity samples. Samples G and H, the low oligosaccharide mutant lines, had the greatest crude protein content and the lowest fiber, NDF, ADF, raffinose, and stachyose content.

Overall, a 10% relative difference in CP was observed between samples (range 48.2 to 53.1%). For fiber this was 115% (range 2.6 to 5.6%), NDF was 57% (range 5.8 to 9.2%), ADF was 112% (range 2.6 to 5.0%), and hemicellulose was 60% (range 3.1 to 4.7%). For raffinose and stachyose, this difference was much larger, 713% (range 0.15 to 1.22%) and 2,018% (range 0.28 to 5.93%), respectively. These data suggest that important differences in composition exist between different varieties of soybean meal, which may have an impact on nutritional value.

Relationships Among Compositional Variables
The crude protein content in these soybean meal samples was negatively correlated with fiber (r = –0.70, P = 0.05), raffinose (r = –0.90, P < 0.01), and stachyose (r = –0.96, P < 0.01) content, in line with findings of Krober and Cartter (1962)Go. Fiber, ADF, and NDF were correlated in these samples (r = 0.86 to 0.92, P < 0.01). This was expected because these are all measures of the fiber content of the samples. Excluding sample G, the phytic acid content varied within a narrow range (0.35 to 0.46%) and was positively correlated with the content of ash (r = 0.92, P < 0.01) because phytate P is a major constituent of ash.

Trypsin inhibitor, urease value, and PDI data suggest that these samples were all processed to a similar extent (C. M. Parsons, Urbana, IL, personal communication), although approaching overprocessed (Herkelman et al., 1992Go; Batal et al., 2000Go; Kim et al., 2001Go). These responses were also correlated with r-values ranging from 0.63 to 0.83 (P = 0.09 and 0.01, respectively).

Ileal Digestibility
Sample H generally had the greatest digestibility (except for Trp), sample B had the lowest digestibility for DM and energy, and samples A and F had low digestibilities for CP, Lys, Thr, and Met (Table 5Go). Among samples, ileal DM digestibility ranged from 79.6 to 83.2%. Energy digestibility was slightly greater (approximately 2 percentage points) than DM digestibility but followed the same trend; their correlation coefficient was 0.91 (P < 0.05). Crude protein digestibility ranged from 80.6 to 84.6%. Digestibility of CP was not correlated with the digestibility of DM (r = 0.53, P = 0.18) or energy (r = 0.45, P = 0.26).


View this table:
[in this window]
[in a new window]
  		Table 5. Apparent ileal digestibility of 8 soybean meal samples (n = 8 observations per mean)
 
The differences in ileal digestibility between the samples with the greatest and lowest digestibility among soybean samples were 5 ± 1% for all variables (Table 5Go) except Thr, for which a difference of 9% was observed. Lower digestibilities for Thr are commonly observed (NRC, 1998Go) and may be related to high endogenous losses of Thr (Stein et al., 1999Go). Relative differences between samples were fairly well preserved within this sample set, suggesting that differences in apparent digestibility uniformly affected endogenous losses.

The apparent ileal digestibility of CP and Lys, Met, and Thr was highly correlated with r > 0.96 (P < 0.01). The high correlation for the Maillard reaction-prone lysine (Assoumani et al., 1993Go) suggests that the Maillard reaction occurred uniformly in the different soybean samples. This also indicates that, practically, processing conditions were similar. Tryptophan digestibility yielded significant correlations with digestibility of CP, Thr, and Met (but not Lys) using classical correlation analysis. However, upon principal component analysis (not shown) Trp showed a different trend than Lys, Met, and Thr similar to what was observed by van Kempen et al. (2002)Go. The reason for this is unknown but may be related to the use of a separate assay for Trp than for other AA (AOAC, 1995Go), resulting in additional variation between Trp and other AA. Protein dispersibility index, trypsin inhibitor, and urease content also had little impact on digestibility except that a negative correlation between both protein dispersibility index and trypsin inhibitor, and apparent ileal DM digestibility was observed (r = –0.69 to –0.73, P = 0.06 to 0.04).

Stachyose content of the test diets used for the ileal digestibility assays was the only compositional variable that had a significant influence on ileal digestibility. Stachyose had a negative correlation with ileal DM (r = –0.80, slope = –0.43 ± 0.13, P = 0.02) and energy (r = –0.73, slope = –0.41 ± 0.16, P = 0.04) digestibility. These slope values suggest that a substantial portion (at least 57%) of the stachyose is degraded in the small intestine, in line with observations of Smiricky et al. (2002)Go. The digestibility of ileal DM (r = –0.74, P = 0.03) and ileal energy (r = –0.72, P = 0.04) was also negatively correlated with the calculated total concentration of hemicellulose and oligosaccharides. However, the r-value obtained with hemicellulose and oligosaccharides was comparable to that obtained with stachyose, indicating that these correlations truly describe the effect of stachyose.

Several poultry experiments have shown that soy oligosaccharides decrease energy digestibility (Coon et al., 1990Go; Leske et al., 1993Go). A possible reason is that oligosaccharide fermentation increases acidity of the lower intestinal tract, thereby increasing digesta passage rate (Coon et al., 1990Go). Another possible reason for flow-rate increase is that soy oligosaccharides are unabsorbable, as are some of the breakdown products, such as melibiose and manninotriose. The presence of unabsorbable sugars and correspondingly retained electrolytes may increase fluid volumes in the small intestine (Wiggins, 1984Go) and lower digestibility.

The correlation of stachyose with ileal CP digestibility was not significant (r = –0.28, P = 0.50). This contrasts data from Smiricky et al. (2003)Go and Houdijk et al. (1998)Go, but these authors used much greater levels of stachyose than what was used in the current study (up to 1.83%). The levels used in this study, however, are in line with what is found in typical corn-soybean meal diets, suggesting that in practical diets, stachyose is not a major factor in protein digestibility.

Total Tract Digestibility
Diets were formulated using data from the ileal digestibility trial for the total tract digestibility assay. Thus, the actual inclusion rate of soybean meal was affected by its nutritional value and ranged from 24 to 28%. Nevertheless, inclusion rate had no effects on total tract digestibility (P > 0.10).

Differences among samples in energy and DM digestibility were smaller at the total tract than at the ileal level (range was 86.5 to 87.7% for total tract energy digestibility, and 87.9 to 88.9% for total tract DM digestibility, Table 6Go). A relative difference in CP digestibility of 4% was observed (range 84.4 to 87.5%), with sample G having the greatest digestibility and sample F the lowest. Large numerical differences in estimated fermentation were observed, with samples C and E apparently loosing less than 4 g of CP/kg in the large intestines due to fermentation, whereas samples A and G apparently lost over 10 g of CP/kg due to fermentation. The lowest estimated DM fermentation was 60 g/kg and the greatest was 74 g/kg. These data were not analyzed statistically because they were estimates.


View this table:
[in this window]
[in a new window]
  		Table 6. Apparent total tract digestibility and estimated crude protein and dry matter fermentation of 5 soybean meal samples formulated into complete feeds (n = 10 observations per mean)
 
Despite the low variation, fecal digestibility of CP, energy, and DM were strongly correlated with each other, with r-values exceeding 0.91 (P = 0.03). These variables were also correlated with calculated CP fermentation (r > 0.87) and with calculated DM fermentation (r > 0.80). Despite respectable correlation coefficients the regression coefficients calculated for CP and DM fermentation were small, around 0.1. Total tract CP digestibility was the exception where a regression coefficient of 0.32 ± 0.07 was assigned to calculated CP fermentation. This is another indicator that fermentation served as an equalizer between diets. In fact, those samples with poorer ileal CP digestibility were more extensively fermented in the large intestine, yielding a greater extent of total tract digestion than samples with high ileal CP digestibility.

Ileal energy digestibility was not correlated with total tract energy digestibility (r = –0.12, P = 0.85). Given that energy digested in the small intestine is presumed to be available for the host, whereas energy released through fermentation in the large intestines is predominantly used by microbes (Wenk, 2001Go), these data suggest that digestibility for both CP and energy should ideally be determined at the ileal level.

Ileal indigestible soybean meal crude protein (prorated for inclusion rate in the test diets) was positively correlated with total tract DM digestibility (r = 0.87, P = 0.06), and calculated DM (r = 0.90, P = 0.04) and CP fermentation (r = 0.89, P = 0.04). These findings suggest that samples with low ileal CP digestibility stimulate fermentative activity in the large intestine, resulting in a more complete degradation of both DM and CP in the large intestine (Figure 1Go).


Figure 1
View larger version (13K):
[in this window]
[in a new window]
  		Figure 1. Relationship between ileal CP digestibility and estimated CP and DM fermentation.

 
Odorants
Numerically large differences in odorant concentration were observed between manure from pigs fed diets with different soybean meal samples (Table 7Go). On average there was a 3-fold greater concentration of odorants between the lowest and the greatest for both fresh and 5-d-old manure. This indicates that the sample of soybean meal used has a major impact on odorant concentration, and thus possibly on odor emission. However, this difference was only statistically significant for a few of the odorants because variation in odorant measurements masked sample effects. Sample G, which was low in stachyose and raffinose but which also had a high calculated CP fermentation, generally had the greatest odorant concentration. A relationship between CP fermentation and odor was expected because odorants are produced in part by the colonic fermentation of endogenous and undigested AA (Tabor and Tabor, 1985Go).


View this table:
[in this window]
[in a new window]
  		Table 7. Odorant concentration in fresh and 5-d-old manure (ppb, n = 10 observations per mean)1
 
Soybean meal-derived components in the complete feed had effects on odorants. Raffinose was positively correlated with indole and phenol (r > 0.91, P < 0.03) and negatively correlated with 4-ethylphenol (r = 0.89, P = 0.04) in fresh manure. Raffinose was negatively correlated with propionate and 3-ethyl phenol in 5-d old manure (r < –0.88, P < 0.05). Stachyose content in the test diet was negatively correlated with propionate, 3-ethylphenol, 4-ethylphenol, and 2-methylindole in 5-d-old manure.

Both soybean meal-derived fiber and the NDF content of the test diet were positively correlated with propionate (r > 0.89, P = 0.04), in line with the expectation that fiber fermentation results in the production of propionate. Such a correlation was not seen with stachyose (r = 0.46, P = 0.44) and raffinose (r = 0.04, P = 0.95) possibly because they are fermented earlier in the intestinal tract, allowing propionate produced to be absorbed (Jørgensen et al., 1996Go). Data from Smiricky et al. (2002)Go also suggest that stachyose is predominantly degraded in the small intestine, whereas nearly two-thirds of the raffinose is degraded in the small intestines. This may also explain the lack of strong effects of stachyose on odor and ammonia emission and the only modest effect of raffinose on odor.

The mercaptan dimethyldisulfide tended to be positively correlated with soybean meal derived crude protein, fiber, NDF, and ADF content in the test diet (r > 0.82, P < 0.09). Estimated CP fermentation was positively correlated with 3- and 4-ethylphenol, tetradecane, and tridecane but negatively correlated with indole and phenol (r < –0.76, P < 0.14) in fresh manure. Estimated DM fermentation was positively correlated with dodecane and tridecane (r > 0.80, P < 0.10) in fresh manure, both degradation products of fatty acids.

Ammonia
Although odorant concentrations were not affected by animal (P > 0.10), ammonia emission was strongly affected by animal (P = 0.03). One of the factors responsible for the animal effect was the ratio of feces to urine produced. On average, manure contained 11.2% feces (with the remainder being urine). Among treatments, this ranged from 10.2% for diets A and G, to 13.0% for diet C (SEM = 1.1). However, among animals the percentage of feces in manure ranged from 7.6 to 18.3% (SEM = 1.6). Percentage of feces was a significant covariate in the statistical analysis of ammonia emission, where a greater percentage of feces in the manure resulted in greater ammonia emission (Figure 2Go). This relationship can be explained by microbial urease in feces promoting the conversion of urinary urea into ammonia. Thus, greater levels of fecal contamination result in faster hydrolysis of urea, causing a more rapid release of ammonia (Tamminga, 1992Go). Ammonia emission could be expressed as a function of time and the feces to manure ratio using the following sigmoidal model (r2 = 0.74, Figure 2Go):


Figure 2
View larger version (63K):
[in this window]
[in a new window]
  		Figure 2. Model (R2 = 0.74) of the relationship between fecal content in manure (ratio), time, and ammonia emission (% of A [asymptote] value in the model describing emission—see text for details). Urine and feces were mixed based on their ratios of production, and in vitro ammonia emission was measured after 12, 24, 36, and 48 h.

 

Formula

in which A = 88 ± 21, b = 1.53 ± 0.15, c = 0.048 ± 0.012, d = 0.82 ± 0.15, and e = 11.5 ± 2.8 with time in hours and ratio as a fraction of feces to manure. This model agrees with the data from Kaspers (2002)Go.

Ammonia emission is predominantly affected by N excreted with urine (Nahm, 2003Go). For this study, diets were formulated to have an equal digestible lysine content using corn and the test soybean meal samples as the sole lysine sources and lysine was the first limiting AA. As a result, the digestible CP content varied only from 14.8 to 15.8%, and total CP varied from 18.1 to 18.7% and dietary CP did not correlate with ammonia emission (P > 0.10; Table 8Go). The only soybean meal-derived component in the complete feed that affected ammonia was raffinose, which tended to be negatively correlated with ammonia emission but only at 48 h (r = –0.81, P = 0.10).


View this table:
[in this window]
[in a new window]
  		Table 8. In vitro ammonia emission (mmol) over time from 500 mL of manure (n = 10 observations per mean)
 

   IMPLICATIONS
Top
 Abstract
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 IMPLICATIONS
LITERATURE CITED
 
Substantial variation in nutritional value between different soybean samples obtained from a genetic database exists, making it possible to select for varieties that are preferred for swine nutrition from both an environmental and nutritional perspective. Low stachyose varieties seem preferable for maximizing energy digestibility. However, no compositional variable was identified that explained differences in ileal crude protein digestibility. Maximizing ileal crude protein digestibility is of interest for nutritional value and may also reduce ammonia and odor emission.


   Footnotes
 
1 Partial funding for this project was provided by the North Carolina Soybean Association, USDA, North Carolina Agricultural Research Service, and Pioneer Hi-Bred International Inc. The use of trade names does not imply endorsement by the North Carolina Agricultural Research Service of the products named or criticism of similar ones not mentioned. Back

3 Current address: Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh 27606. Back

4 Pioneer—A DuPont Company, Des Moines, IA, current address: Kemin Industries, 2100 Maury Street, Des Moines, IA 50317. Back

2 Correspondence and current address: Provimi Research & Technology Centre, Lenneke Marelaan 2, B-1932 St. Stevens Woluwe, Belgium (theovankempen{at}yahoo.com).

Received for publication May 14, 2005. Accepted for publication January 14, 2006.


   LITERATURE CITED
Top
 Abstract
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 IMPLICATIONS
 LITERATURE CITED
 


AOAC. 1995. Official Method of Analysis. 16th ed. Assoc. Off. Anal. Chem., Arlington, VA.

AOCS. 1998. Official Methods and Recommended Practices of the AOCS. 5th ed. D. Firestone, ed. Am. Oil Chem. Soc., Champaign, IL.

Assoumani, M. B., D. Maxime, and N. P. Nguyen. 1993. Evaluation of a lysine-glucose Maillard model system using three rapid analytical methods. Pages 43–50 in Maillard Reactions in Chemistry, Food, and Health. T. P. Labuza, G. A. Reineccius, V. M. Monnier, J. O’Brien, and J. W. Baynes, ed. Royal Soc. Chem., Cambridge, UK.

Batal, A. B., M. W. Douglas, A. E. Engram, and C. M. Parsons. 2000. Protein dispersibility index as an indicator of adequately processed soybean meal. Poult. Sci. 79:1592–1596.[Abstract/Free Full Text]

Coon, C. N., O. Akavanichan, and T. K. Cheng. 1990. Effect of oligosaccharide-free soybean meal on true metabolizable energy and fiber digestion in adult roosters. Poult. Sci. 69:787–793.[Medline]

Cromwell, G. L. 2000. Utilization of soy products in swine diets. Pages 258–282 in Soy in Animal Nutrition. J. K. Drackley, ed. Fed. Anim. Sci. Soc., Savoy, Illinois, USA.

Gralapp, A. K., W. J. Powers, and D. W. Bundy. 2001. Comparison of olfactometry, gas chromatography, and electronic nose technology for measurement of indoor air from swine facilities. Trans. ASAE 44:1283–1290.

Gralapp, A. K., W. J. Powers, M. A. Faust, and D. S. Bundy. 2002. Effects of dietary ingredients on manure characteristics and odorous emissions from swine. J. Anim. Sci. 80:1512–1519.[Abstract/Free Full Text]

Herkelman, K. L., G. L. Cromwell, T. S. Stahly, T. W. Pfeiffer, and D. A. Knabe. 1992. Apparent digestibility of amino acids in raw and heated conventional and low-trypsin-inhibitor soybeans for pigs. J. Anim. Sci. 70:818–826.[Abstract]

Hitz, W. D., T. J. Carlson, P. S. Kerr, and S. A. Sebastian. 2002. Biochemical and molecular characterization of a mutation that confers a decreased raffinosaccharide and phytic acid phenotype on soybean seeds. Plant Phys. 128:650–660.[Abstract/Free Full Text]

Houdijk, J. M. G., B. A. Williams, S. Tamminga, and M. W. A. Verstegen. 1998. Effect of non-digestible oligosaccharides in the diet for weaner pigs on in vitro fermentation. Page 30 in Proc. Brit. Soc. Anim. Sci. BSAS, Penicuik, UK.

Jørgensen, H., X. Q. Zhao, K. E. Bach Knudsen, and B. Eggum. 1996. The influence of dietary fiber source and level on the development of the gastrointestinal tract, digestibility and energy metabolism in broiler chickens. Br. J. Nutr. 75:379–395.[Medline]

Kaspers, B. A. 2002. Designing a conveyor belt waste collection system for swine. M.Sc. Thesis, North Carolina State Univ., Raleigh.

Kim, S. W., Z. H. Zhang, L. A. Johnson, and R. A. Easter. 2001. Apparent and true ileal digestibility of amino acids in soybean meals as affected by heat treatments and trypsin inhibitors. J. Anim. Sci. 79(Suppl. 1):321–322. (Abstr.)

Krober, O. A., and J. L. Cartter. 1962. Quantitative interrelationships of protein and non-protein constituents of soybeans. Crop Sci. 2:171–172.[Free Full Text]

Leske, K. L., C. J. Jevne, and C. N. Coon. 1993. Effect of oligosaccharide additions on nitrogen-corrected true metabolizable energy of soy protein concentrate. Poult. Sci. 72:664–668.[Medline]

Moeser, A. J., I. B. Kim, E. van Heugten, and T. A. T. G. van Kempen. 2002. The nutritional value of degermed, dehulled corn for pigs and its impact on the gastrointestinal tract and environmental pollution. J. Anim. Sci. 80:2629–2638.[Abstract/Free Full Text]

Nahm, K. H. 2003. Influence of fermentable carbohydrates on shifting nitrogen excretion and reducing ammonia emission of pigs. Crit. Rev. Envir. Sci. Techn. 30:165–186.

NRC. 1998. Nutrient requirements of swine. 10th ed. Natl. Acad. Press, Washington, DC.

Palmer, R. G., T. Hymowitz, and R. L. Nelson. 1996. Germplasm diversity within soybean. Pages 1–37 in Soybean: Genetics, Molecular Biology and Biotechnology. D. P. S. Verma and R. C. Shoemaker, ed. CAB Int., Guildford, UK.

Smiricky, M. R., C. M. Grieshop, D. M. Albin, J. E. Wubben, V. M. Gabert, and G. C. Fahey Jr. 2002. The influence of soy oligosaccharides on apparent and true ileal amino acid digestibilities and fecal consistency in growing pigs. J. Anim. Sci. 80:2433–2441.[Abstract/Free Full Text]

Smiricky, M. R., C. M. Grieshop, E. A. Flickinger, L. L. Bauer, and G. C. Fahey Jr. 2003. Dietary galactooligosaccharides affect ileal and total-tract nutrient digestibility, ileal and fecal concentrations, and ileal fermentative characteristics of growing pigs. J. Anim. Sci. 81:2535–2545.[Abstract/Free Full Text]

Stein, H. H., N. L. Trottier, C. Bellaver, and R. A. Easter. 1999. The effect of feeding level and physiological status on total flow and amino acid composition of endogenous protein at the distal ileum in swine. J. Anim. Sci. 80:1180–1187.

Suna, K., K. Wansoo, and I. K. Han. 2003. Optimization of the extraction and purification of oligosaccharides from defatted soybean meal. 38. Int. J. Food Sci. Techn. 3:337–342.

Tabor, C. W., and H. Tabor. 1985. Polyamines in microorganisms. Microbiol. Rev. 49:81–99.[Free Full Text]

Tamminga, S. 1992. Gaseous pollutants produced by farm animal enterprises. Pages 345–357 in Farm Animals and the Environment. C. Phillips and D. Piggins, ed. CAB Int,, Wallingford, UK.

van Kempen, T. A., I. B. Kim, A. J. Jansman, M. W. Verstegen, J. D. Hancock, D. J. Lee, V. M. Gabert, D. M. Albin, G. C. Fahey, C. M. Grieshop, and D. Mahan. 2002. Regional and processor variation in the ileal digestible amino acid content of soybean meals measured in growing swine. J. Anim. Sci. 80:429–439.[Abstract/Free Full Text]

Wenk, C. 2001. The role of dietary fibre in the digestive physiology of the pig. Anim. Feed Sci. Tech. 90:21–33.

Wiggins, H. S. 1984. Nutritional value of sugars and related compounds undigested in the small gut. Proc. Nutr. Soc. 43:69–75.[Medline]

Willis, R., M. Montgomery, and P. Allen. 1996. Improved method for manual, colorimetric determination of total Kjeldahl nitrogen using salicylate. J. Agric. Food Chem. 44:1804–1807.



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by van Kempen, T. A. T. G.
Right arrow Articles by Sewalt, V. J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van Kempen, T. A. T. G.
Right arrow Articles by Sewalt, V. J. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS