|
|
||||||||

* Prairie Swine Centre Inc., Saskatoon, SK, Canada S7H 5N9 and
and
Department of Animal and Poultry Science, University of Saskatchewan, Saskatoon, Canada S7N 5A8
3 Correspondence:
P.O. Box 21057, 2105 8th St. E. (phone: 306-373-9922; fax: 306-955-2510; E-mail:
ruurd{at}sask.usask.ca).
| Abstract |
|---|
|
|
|---|
Key Words: Excretion Fibers Nitrogen Pigs Protein Urea
| Introduction |
|---|
|
|
|---|
Urinary N is related to plasma urea (Brown and Cline, 1974). Plasma urea at 4 h postfeeding predicted daily urinary N excretion best for grower pigs with restricted access to feed (Zervas and Zijlstra, 2002), but models for pigs with free access to feed have not been developed.
The objectives of this study were 1) to determine if the two manipulations of reduced dietary protein and fermentable fiber would together reduce urinary N excretion further than a single manipulation, 2) to determine if effects of diet manipulations were similar between pigs with restricted and free access of feed, 3) to determine if fermentable fiber sources reduce voluntary feed intake, and 4) to further develop predictions of urinary N excretion using plasma urea.
| Materials and Methods |
|---|
|
|
|---|
|
|
During collections, representative feed samples were collected. Feces were collected twice daily, pooled, and stored at -20°C until analyses. Urine was collected twice daily, weighed, and a 5% aliquot was stored at -20°C. Twenty milliliters of 12 N HCl was added to the collection container prior to each collection to prevent volatilization of urinary N. Blood samples were centrifuged and plasma was frozen at -20°C. After the collection, feces and urine samples were thawed, homogenized, and subsampled, and feces were freeze-dried.
Chemical Analyses
Feed and freeze-dried fecal samples were ground through a 1-mm screen in a Retsch mill (Brinkman Instruments, Rexdale, ON, Canada). Chemical analyses were conducted in duplicate. Feed, feces, and urine samples were analyzed for N by combustion (method 968.06; AOAC, 1990) using a Leco protein/nitrogen determinator (model FP-528, Leco Co., St Joseph, MI). Dry matter content of feed and feces was determined by drying at 135°C in an airflow-type oven for 2 h (method 930.15; AOAC, 1990). Chromic oxide content was analyzed in feed and feces (Fenton and Fenton, 1979) with a Pharmacia LKB-Ultrospec III spectrophotometer (model 80-2097-62; Cambridge, England), at 440 nm after ashing at 450°C overnight. Gross energy in feed and feces was measured in an adiabatic bomb calorimeter (model 1281, Parr Instrument Co., Moline, IL). The ADF and NDF contents were determined using an Ankom200 fiber analyzer (Ankom Technology Co., Fairport, NY). Feed samples were analyzed for AA (method 994.12; AOAC, 1995). Methionine was determined as methionine sulfone and cystine as cysteic acid after oxidation with performic acid. Tryptophan was determined after alkaline hydrolysis with lithium hydroxide by means of reversed phase high performance liquid chromatography. Plasma urea was analyzed using the Abbott Spectrum urea nitrogen test (Series II, Abbot Laboratories, Dallas, TX). Apparent total-tract digestibility of N and energy, N retention, and DE were calculated using chromic oxide concentrations in diets and feces, using the indicator method.
Statistical Analyses
The individual pig was considered the experimental unit. Variables were analyzed using the GLM procedure of SAS (SAS Inst., Inc., Cary, NC). The statistical model included effects for dietary treatment (protein, fermentable fiber, and protein x fiber interaction) and group; initial BW was included as covariate. Means comparisons were performed using the probability of difference. Regression analysis was used to predict urinary N as a function of plasma urea. Values are reported as least squares means.
| Results |
|---|
|
|
|---|
Diets were mixed in two batches of 200 kg. The analyzed protein content was low for one of two batches for two diets (high- and low-protein SH); part of the soybean meal was probably not included during diet mixing. Thus, observations from pigs fed these two batches were not included in statistical analysis for the N balance study, but the plasma urea and urinary N values were included in regression analysis to investigate the relation between plasma urea and urinary N excretion.
Nitrogen Balance.
Nitrogen intake was reduced 15% for low-protein compared to high-protein diets (Table 3
; P < 0.001), differed among fiber diets (P < 0.001), and was affected by a protein x fiber interaction (P < 0.01). The interaction was due to a 0.6 or 1.1% higher protein content in control diets and a 0.5% lower protein content in the high-protein SH diet, compared to calculated protein values, indicating the importance of ingredient analyses before diet mixing.
|
Expressed as a percentage of N intake, total N excretion and retention (N utilization) were not affected by dietary protein (P > 0.10). Nitrogen intake was 4% lower for SH and SBP compared to control diets (Table 3
; P < 0.001), and differences for N excretion patterns and N retention (expressed in g/d) may therefore be partly due to differences in N intake. Effects of fermentable fiber on fecal and urinary N excretion patterns are therefore compared best as percentages of N intake in the present study. Fermentable fiber increased fecal and reduced urinary N excretion (P < 0.01), but did not affect total N excretion or N retention (P > 0.10). Fecal N was increased 4% units for SH and 6.5% units for SBP compared to control diets (P < 0.01). Urinary N was reduced 5% units for SH (P < 0.10) and 9% units for SBP (P < 0.05) compared to control diets.
Energy and Nitrogen Digestibility.
The DE content was affected by fiber (Table 3
; P < 0.001) and a protein x fiber interaction (P < 0.05). The DE content was 2 to 4% lower for SBP and control diets, but 2% higher for the high-protein SH diet compared to the calculated DE content. The measured DE content was 0.10 Mcal/kg higher for SH compared to control diets (P < 0.001). Energy digestibility was affected by fiber (P < 0.001), but not by protein (P > 0.10). Digestibility of energy was 2% units lower for SH and SBP compared to control diets (P < 0.01). The N digestibility was reduced 3% units for low- compared to high-protein diets (P < 0.001), and was reduced 4% units for SH and 6.5% units for SBP compared to control diets (P < 0.01).
Performance.
Feed efficiency and ADG were affected by fiber (Table 3
; P < 0.10), but not protein (P > 0.10); performance was 9% higher for SBP compared to control diets (P < 0.05).
Plasma Urea.
Dietary protein and fiber affected plasma urea (Table 3
; P < 0.01). Preprandial plasma urea was reduced 29% for low- compared to high-protein diets (P < 0.01) and was reduced 17% for SH (P < 0.10) and 39% for SBP (P < 0.001) compared to control diets. At 4 h postfeeding, plasma urea was reduced 32% for low- compared to high-protein diets (P < 0.001), and reduced 15% for SH (P < 0.10) and 26% for SBP (P < 0.001) compared to control diets. Compared to preprandial levels, plasma urea at 4 h postfeeding was increased 38% for the high-protein diet, 33% for the low-protein diet, 28% for SH, 30% for SBP, and 54% for control diets. Plasma urea at 4 h postfeeding was related better to daily urinary N excretion (R2 = 0.66) than preprandial plasma urea (R2 = 0.47). At 4 h postfeeding, urinary N excretion (g/d) was predicted by 1.93 + 2.06 x plasma urea concentration (mmol/L). The residual standard deviation (RSD) was 1.74 g/d and the 95% prediction interval ranged from 8.8 to 16.1 g/d for the average plasma urea concentration of 5.0 mmol/L.
Free Access to Feed
One pig with a rectal prolapse was removed from the study. A large volume of urine (contamination with water) prevented accurate estimation of the urinary N content for another pig; this pig was not used in statistical analyses.
Nitrogen Balance.
Nitrogen intake was reduced 12% for the low- compared to high-protein diets (Table 4
; P < 0.01) and differed among fiber diets (P < 0.05). Expressed in grams per day, urinary N excretion was reduced 27% (P < 0.05), total N excretion was reduced 16%, N retention was decreased 7% (P < 0.10), and fecal N was not different (P > 0.10) for low-protein compared to high-protein diets. Fiber affected urinary N excretion (P < 0.01), but not total N excretion or N retention (P > 0.10). None of the N variables was affected by a protein x fiber interaction (P > 0.10). The ratio of urinary to fecal N was reduced 28% for low-protein compared to high-protein diets (P < 0.01) and was reduced 43% for SH and 55% for SBP compared to control diets (P < 0.001).
|
Energy and Nitrogen Digestibility.
The measured DE content was 3% lower with free compared to restricted access to feed. The DE content was affected by fiber (P < 0.001) and a protein x fiber interaction (P < 0.01). The DE content was 0.10 Mcal/kg higher for SBP (P < 0.001) and 0.05 Mcal/kg higher for SH (P < 0.10) compared to control diets. Energy digestibility was affected by fiber (Table 4
; P < 0.001), but not by protein (P > 0.10). Digestibility of energy was reduced 3% units for SH and 4% units for SBP compared to control diets (P < 0.001). The N digestibility was reduced 2.5% units for the low- vs high-protein diets (P < 0.01), 4% units for SH, and 6.5% units for SBP compared to control diets (P < 0.01).
Animal Performance.
Feed intake and efficiency were not affected by protein or fiber (Table 4
; P > 0.10), but ADG was reduced 11% for SH compared to control diet (P < 0.05), in part by an 8% lower feed intake for SH diet (P < 0.10).
Plasma Urea.
Dietary protein and fiber affected plasma urea (Table 4
; P < 0.10). At 0800, plasma urea was 35% higher for the high- vs low-protein diets (P < 0.05), and plasma urea was reduced 28% for SH and 21% for SBP compared to control diets (P < 0.05). At 1200, plasma urea was 29% higher for the high vs low-protein diets (P < 0.10), and plasma urea was reduced 32% for SH (P < 0.05) and 18% for SBP (P < 0.10) compared to control diets. Daily urinary N excretion could be predicted by plasma urea at 0800 (Figure 1
; R2 = 0.71) and at 1200 (R2 = 0.65). At 0800, urinary N excretion (g/d) was predicted by -0.62 + 3.65 x plasma urea concentration (mmol/L). The RSD was 3.78 g/d, and the 95% prediction interval ranged from 10.1 to 25.9 g/d for the average plasma urea concentration of 5.0 mmol/L.
|
| Discussion |
|---|
|
|
|---|
Dietary Protein
The reduction in urinary and total N excretion was similar to Gatel and Grosjean (1992), Dourmad et al. (1993), and Canh et al. (1998a). Similar to Canh et al. (1998a), fecal N excretion was not reduced because the reduced N intake was compensated by a reduction in N digestibility for the low- vs high-protein diets, resulting in similar fecal N excretion. With restricted access to feed, for each percent reduction in dietary protein, urinary N excretion was reduced 10% (1.2 g/d) and total N excretion was reduced 4% (1.2 g/d), which is less compared to the lowest reductions for urinary N (8% or 1.9 g/d) and total N excretion (8% or 2 g/d) reported in other studies (Gatel and Grosjean, 1992; Dourmad et al., 1993; Canh et al., 1998a). With free access to feed, for each percent reduction in dietary protein, urinary N excretion was reduced 10% (1.8 g/d) and total N excretion was reduced 6% (1.8 g/d). To compare studies properly, reductions in N excretion should be expressed as a function of feed intake. Ideally, each percentage of reduction in dietary protein will result in 1.6 g less urinary and total N excreted per kilogram of feed consumed (10 g protein = 1.6 g N), if N retention and fecal N excretion are not affected. Urinary N was reduced 0.8 to 1 g and total N was reduced 1 to 1.3 g for control and SBP diets for each percentage of reduction in dietary protein per kg of feed consumed, whereas reductions ranging from 0.9 to 1.9 g in urinary or total N excretion (Gatel and Grosjean, 1992; Canh et al. 1998a) have been reported. Dissimilarities among studies can be attributed to differences in N retention between pigs fed high- and low-protein diets, or perhaps an inability to trap all N in excreted urine. Reducing dietary protein reduced the urinary:fecal N excretion ratio. Because fecal N was not affected by dietary protein, the reduced ratio was caused solely by the reduction in urinary N excretion.
With restricted access to feed, N retention was reduced 4% (0.9 g/d) for each percentage of reduction in dietary protein. A reduction in N retention may be attributed to differences in intake of essential AA, especially lysine (e.g., resulting from an analyzed lysine content that is lower than calculated) or to a lower utilization of synthetic AA in pigs that are fed infrequently (Batterham et al., 1984; Partridge et al., 1985). However, the theory of reduced efficiency of utilization of synthetic AA in low-protein diets due to a discrepancy of absorption between synthetic AA and AA from protein sources was disputed recently for pigs fed at least two meals per day (Le Bellego et al., 2001), and may thus not be valid for the conditions of the present study. Indeed, with free access to feed, N retention remained reduced for low-protein diets indicating that intake of AA caused the reduction in N retention and not infrequent feeding.
Dietary protein did not affect feed efficiency and ADG, despite reductions in N retention. This can be attributed partly to changes in fat deposition because of energy sparing with low-protein diets (Noblet et al., 1987). As expected, apparent total-tract N digestibility was reduced with low-protein diets in the present study, because of changing ingredient ratios. Calculated using average CP content of ingredients, soybean meal provided 44% of N in high-protein diets and 31% of N in low-protein diets, thereby reducing the relative amount of highly digestible protein in low-protein diets.
Fermentable Fiber
In the present study, SH and SBP shifted N excretion from urine to feces. Generally, fiber that escapes digestion and absorption in the small intestine is fermented in the large intestine into VFA. Protein that escapes digestion and absorption as AA in the small intestine is converted partly in the large intestine into ammonia, which is absorbed into blood, transformed to urea in the liver, and excreted as urea in urine (Mosenthin et al., 1992). Microflora in the large intestine will use fiber as energy and ammonia as N for protein synthesis, thereby reducing ammonia absorption into blood and urea excretion in urine. Microbial protein is excreted in feces and fermentable fiber will therefore shift N excretion from urinary to fecal N (Morgan and Whittemore, 1988; Canh et al., 1997). To reduce ammonia emission, fecal N is preferable to urinary N because fecal N (mainly bacterial protein) is less susceptible to rapid decomposition than urinary N (mainly urea), which is easily converted into ammonium and carbon dioxide by fecal bacterial urease (Canh et al., 1998b; Mroz et al., 2000). In manure, ammonium is in equilibrium with ammonia, and ammonia emission is reduced by reduced ammonium concentration and reduced pH (Aarnink, 1997; Stevens et al., 1989). Apart from affecting N excretion patterns, fermentable fiber will increase VFA concentration in feces and manure and thereby reduce manure pH (Canh et al., 1998b). Fermentable fiber may reduce ammonium concentration and pH of the manure, and will thus reduce ammonia emission from manure for a combination of factors (Canh et al., 1998c).
Soyhulls and SBP reduced urinary N and increased fecal N excretion, without affecting N retention and total N excretion, similar to the findings of Canh et al. (1997). Urinary N excretion of grower pigs was reduced 26 and 31% with 20% SBP in the diet and 15 and 26% with 15% SH in the diet, with restricted and free access to feed, respectively. Urinary N excretion of finisher pigs was reduced 44% with 30% SBP in the diet (Canh et al., 1997). Fecal N increased with 21% SBP and 16% SH in the diet compared to 28% tapioca (Mroz et al., 2000), but without a reduction in urinary N. The lack of reduction was caused partially by an imbalance in AA in the SBP diet, reflected by a reduced N retention.
In the present study, the ratio of urinary to fecal N excretion was lowest for SBP, intermediate for SH, and highest for control diets. The SBP was not compared directly to SH because SH was included at a lower rate to prevent differences in voluntary feed intake. The 20% SBP clearly affected N excretion patterns, but effects of 15% SH were less profound and appeared less consistent. For the high-protein diets, SH did not affect fecal N, but reduced urinary N compared to the control. For the low-protein diet, SH increased fecal N, but did not affect urinary N compared to the control with restricted access to feed. The lack of reduction in urinary N for the low-protein SH diet may have been caused by an AA imbalance, which was validated by a lower N retention.
The DE content of SH was higher than that of the control diets, indicating that SH diets were overformulated for DE. The DE content of SH was probably underestimated, providing further proof for the limitations of knowledge of energy value for byproducts. Digestibility of N and energy was reduced with SBP and SH, similar to the findings of Chabeauti et al. (1991). Fermentable fiber affects total-tract and ileal digestibility of nutrients (Mroz et al., 2000). The SH and SBP reduced total-tract digestibility of N by 7 and 5% units, respectively, compared to tapioca diets, whereas total-tract digestibility of energy was reduced 2% units for SH. Large intestine fermentation contributes substantially to the digestion of SBP and SH, based on differences between ileal and total-tract digestibility (Mroz et al., 2000). The SBP and SH contain soluble fiber as pectin (Chabeauti et al., 1991), which is fermented easily in the large intestine (Stanogias and Pearce, 1985).
Low-Protein Diets with Fermentable Fiber
Dietary protein and fiber did not interact for N excretion variables or N retention with the exception of fecal N excretion with restricted access to feed; therefore, the effects are additive. Combining the two nutritional strategies will reduce urinary N excretion further than a single strategy without affecting N retention. For example, with free access to feed, the low-protein SBP diet reduced urinary N excretion 55% compared to the high-protein control diet without affecting N retention or ADG; low-protein vs high-protein and SBP vs control reduced urinary N excretion by 27% and 36%, respectively. The additive effects are supported by ammonia emission research (Kreuzer et al., 1998); lowering dietary protein enhanced reductions in emission by fermentable fiber, and the combination reduced ammonia emission 38%. For urinary N excretion, the additive effect was observed for the low-protein SH diet, but ADG and N retention were reduced as well. Generally, changes in plasma urea concentration coincided with changes in urinary N excretion, and plasma urea concentrations further confirmed the additive effects.
Restricted vs Free Access to Feed
Effects of dietary treatments on N excretion patterns were similar between restricted and free access to feed; predictably, N retention was higher for pigs with free access to feed. The chronology of feed allowance is a model for compensatory growth (e.g., Critser et al., 1995) because pigs had free access to feed after a period of nutrient restriction. Specific components of growth were not separated in the present study, but gut fill and the size of visceral organs (to process the increased amounts of nutrients consumed) probably increased rapidly after realimentation (Critser et al., 1995), whereas carcass lean gain may have been less affected. For low- vs high-protein diets, N retention was reduced more with restricted than free access to feed, indicating that the effectiveness of low-protein diets may be underrated if assessed in pigs with restricted access to feed. With restricted access to feed, ADG was higher with the SBP diet than with the control, which may be partially caused by changes in gut fill. During adaptation, pigs were fed a high-fiber diet that probably caused a high level of gut fill at the start of the experiment, which was maintained for the SBP diet, but not for the low-fiber control diet after the switch to experimental diets. With free access to feed, performance was maintained for SBP, but reduced for SH. For SH, either the high DE content or SH fiber reduced voluntary feed intake, thereby reducing ADG. The reduced ADG indicates the importance of verification of treatment effects using grower pigs with free access to feed if the treatments include factors, such as fiber, that reduce voluntary feed intake (Kyriazakis and Emmans, 1995). Digestibility of energy and N was reduced as feed intake increased, probably because the passage rate of digesta was increased, thereby reducing nutrient digestibility (Roth and Kirchgessner, 1985).
Plasma Urea and Urinary Nitrogen
Urea excreted in urine is the main nitrogenous end-product from AA catabolism in pigs, and plasma urea concentrations may be indicative of excreted N in urine (Brown and Cline, 1974; Zervas and Zijlstra 2002). Lowering dietary protein reduced plasma urea with both feeding regimes, similar to findings from Lopez et al. (1994) and Lenis et al. (1999), indicating differences in protein quality. The SH and SBP diets reduced plasma urea, similar to reduced plasma urea and urinary N excretion in rats fed fermentable fiber (Younes et al., 1993). For both dietary treatments, plasma urea reflected urinary N excretion, indicating a relationship between the two variables.
Previously, plasma urea concentrations at 4 h postfeedingl predicted daily urinary N excretion best (Zervas and Zijlstra, 2002), similar to a better prediction for plasma urea at 4 h postfeeding than before-morning feeding in the present study. In pigs with free access to feed, a similar R2 between plasma urea and daily urinary N excretion was obtained for two sampling times 4 h apart, indicating that specific time of blood sampling is less relevant for pigs with free access to feed. The slope of the regression was steeper for pigs with free access to feed than pigs with restricted access (3.65 vs 2.06), but the range in measured plasma urea concentrations among pigs did not differ much between the two feed regimes, indicating that a developed model is only valid for use within a specific feed regime. However, although plasma urea could be used to detect differences among dietary treatments, the 95% prediction intervals of the models indicate that the developed models may not predict urinary N excretion accurately.
| Implications |
|---|
|
|
|---|
| Footnotes |
|---|
2 Supported by project funding from Agriculture and Agri-Food Canada/Natural Sciences and Engineering Research Council of Canada-Research Partnership Program and the Alberta Agriculture Research Institute, and by program funding from Saskatchewan Agriculture
Food and the pork producers of Saskatchewan, Manitoba, and Alberta. S. Zervas received a scholarship from the State Scholarship Foundation of Greece. The authors acknowledge Degussa AG for amino acid assays. ![]()
Received for publication August 21, 2001. Accepted for publication August 6, 2002.
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. M. Hill, H. G. Bateman II, J. M. Aldrich, and R. L. Schlotterbeck Effects of Feeding Different Carbohydrate Sources and Amounts to Young Calves J Dairy Sci, August 1, 2008; 91(8): 3128 - 3137. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. K. Htoo, B. A. Araiza, W. C. Sauer, M. Rademacher, Y. Zhang, M. Cervantes, and R. T. Zijlstra Effect of dietary protein content on ileal amino acid digestibility, growth performance, and formation of microbial metabolites in ileal and cecal digesta of early-weaned pigs J Anim Sci, December 1, 2007; 85(12): 3303 - 3312. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. K. Htoo, W. C. Sauer, Y. Zhang, M. Cervantes, S. F. Liao, B. A. Araiza, A. Morales, and N. Torrentera The effect of feeding low-phytate barley-soybean meal diets differing in protein content to growing pigs on the excretion of phosphorus and nitrogen J Anim Sci, March 1, 2007; 85(3): 700 - 705. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. M. Panetta, W. J. Powers, H. Xin, B. J. Kerr, and K. J. Stalder Nitrogen excretion and ammonia emissions from pigs fed modified diets. J. Environ. Qual., July 1, 2006; 35(4): 1297 - 1308. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Guay, S. M. Donovan, and N. L. Trottier Biochemical and morphological developments are partially impaired in intestinal mucosa from growing pigs fed reduced-protein diets supplemented with crystalline amino acids J Anim Sci, July 1, 2006; 84(7): 1749 - 1760. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. A. Kohn, M. M. Dinneen, and E. Russek-Cohen Using blood urea nitrogen to predict nitrogen excretion and efficiency of nitrogen utilization in cattle, sheep, goats, horses, pigs, and rats J Anim Sci, April 1, 2005; 83(4): 879 - 889. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Coverdale, J. A. Moore, H. D. Tyler, and P. A. Miller-Auwerda Soybean hulls as an alternative feed for horses J Anim Sci, June 1, 2004; 82(6): 1663 - 1668. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. R. Otto, M. Yokoyama, P. K. Ku, N. K. Ames, and N. L. Trottier Nitrogen balance and ileal amino acid digestibility in growing pigs fed diets reduced in protein concentration J Anim Sci, July 1, 2003; 81(7): 1743 - 1753. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Zervas and R. T. Zijlstra Effects of dietary protein and oathull fiber on nitrogen excretion patterns and postprandial plasma urea profiles in grower pigs J Anim Sci, December 1, 2002; 80(12): 3238 - 3246. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |