| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
Department of Animal Sciences, Purdue University, West Lafayette, IN 47097-2054
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
Key Words: endogenous • phosphorus • pig • soybean meal • true digestibility
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
). Swine produce insufficient endogenous phytase to liberate substantial quantities of P from phytate (Pointillart et al., 1984). Although diets formulated with grain and soybean meal (SBM) contain adequate total P, low availability of P necessitates the inclusion of inorganic mineral sources to meet the P requirement of the pig. However, apparent P digestibility estimates do not account for endogenous P loss (EPL) from the animal. By quantifying EPL resulting from feeding individual feedstuffs, nutritionists may better match P supply with animal requirements. In turn, this strategy may reduce the dependence upon inorganic mineral sources, which add significant cost to swine diets and potentially have negative environmental impacts.
Quantification of EPL in swine is difficult due to the inherent shortcomings of current methodologies. For instance, EPL is known to be highly variable, and is influenced by diet composition, mineral concentration, feeding level, energy supply, age, and growth rate of the pig (Jongbloed, 1987). Fan et al. (2001) proposed an approach to quantify EPL by regressing total P output from the animal against dietary P intake. Extrapolation to zero P intake, therefore, provides a theoretical estimate of diet-independent EPL. However, recent studies (Petersen and Stein, 2004; Pettey et al., 2004) have provided EPL estimates that are lower than previous estimates.
The objectives of this study were to: 1) reevaluate regression analysis in estimation of true P digestibility (TPD) and EPL; and 2) compare TPD and EPL of conventional and low-phytate soybean meals. We reasoned that comparison of 2 ingredients, differing mainly in their phytate P content, would provide insight into the influence of dietary phytate content on TPD and EPL in pigs.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
). Thus, dietary Ca:total P ratios were directly related to the Ca:total P ratio of each SBM variety. The Ca:total P ratios were calculated to be static within diets formulated with the same SBM, and all ratios were less than 1:1.
The analyzed nutrient profiles of the conventional and low-phytate SBM varieties are presented in Table 1 and were similar with the exception of P. The nutrient concentrations of the conventional SBM were similar to values summarized by the NRC (1998), with the exception of nonphytate P (i.e., total P minus phytate P). Conventional SBM had a calculated nonphytate P content of 2.8 g/kg compared with an NRC (1998) value of 1.6 g/kg (assuming 23% bioavailability). Using the analyzed phytate P content of each SBM, the diets formulated with the conventional SBM were calculated to contain 0.5, 1.0, 1.5, and 2.0 g of phytate P/kg of diet, whereas the diets formulated with the low-phytate SBM were calculated to contain 0.2, 0.4, 0.6, and 0.8 g of phytate P/kg of diet.
|
) were semipurified, consisting primarily of cornstarch, sucrose, dextrose, and SBM, and were formulated to be similar in caloric density by adjusting the soybean oil. The feed allowance was based on individual metabolic BW of the pigs, as determined at the beginning of each feeding period, and was calculated to be 90 g/kg of BW0.75. One-half of this amount was fed twice daily (0800 and 2000), and the calculated feed intake was static within any given feeding period. To calculate nutrient digestibilities, chromium sesquioxide was incorporated into the diets (3 g/kg, as-fed) as an indigestible marker.
|
). After a 10-d recovery from surgery, pigs were randomly assigned to metabolism crates (1.22 x 1.22 m) equipped with stainless-steel feeders and low-pressure, automatic water nipples.
The experimental diets in meal form were fed to the 16 pigs according to a replicated 8 x 8 standard Latin square design, with each period lasting 7 d. Days 1 through 5 were used for diet acclimation, with fresh fecal grab samples collected on d 5. An adjustment period of 5 d has been shown by our laboratory and by Düngelhoef et al. (1994) to be sufficient to attain steady-state conditions for P metabolism in pigs at this age and, therefore, to minimize dietary carryover effects.
To preempt the possibility of such carryover effects, the diets were rotated via a standard Latin square such that pigs received increasing sequential concentrations of dietary P during each progressive period, regardless of which diet each pig initially received. Therefore, a decrease in P concentration occurred a maximum of twice for each pig (i.e., when rotated to a diet based on a different SBM). This diet rotation minimized the intake of P above the concentrations received in the preceding feeding period for each pig.
Prececal digesta were collected for 12 h on d 6 and 7 by attaching a polyethylene tubular bag (National Bag Company, Inc., Hudson, OH) to the externalized T-cannula. The bags contained 10 mL of 5% formic acid to reduce microbial proliferation, and prececal contents were stored at –20°C between collections. Prececal digesta samples were thawed, pooled for each pig for the 2-d collection, subsampled, and lyophilized. Fecal grab samples collected from each pig were also mixed and subsampled before lyophilization.
To facilitate chemical analyses, the ingredients and diets, as well as freeze-dried prececal digesta and feces, were ground through a 0.75-mm sieve using a grinding mill (Retsch, Inc., Newtown, PA). Dry matter content was determined by drying samples at 105°C for 24 h. Dietary phytate P content was analyzed by Sciantec Analytical Services Ltd. (Thirsk, North Yorkshire, UK) based on the method described by McCance and Widdowson (1935). Chromium, Ca, and P concentrations were determined in triplicate by the inductively coupled plasma atomic emission spectroscopy method (AOAC, 2000; method 990.08) using an Optima model 4300 DV ICP-OES (Perkin Elmer, Boston, MA) after nitric/perchloric acid wet-ash digestion (AOAC, 2000; method 935.13). Phosphorus was determined using axially oriented plasma at a wavelength of 213.617 nm to avoid interference from mineral matrix effects as suggested by the manufacturer; Cr and Ca used plasma in the radial orientation with wavelengths of 267.716 and 317.933 nm, respectively.
The analyzed element concentrations were standardized for recovery of elements from reagent-grade internal quality controls for each analytical run using Cr sesquioxide for Cr and monobasic Ca phosphate for Ca and P. These compounds were chosen as they are commonly used in nonruminant research diets. Recoveries by this methodology were 91 to 97% for Cr, 89 to 99% for Ca, and 99 to 110% for P. Therefore, with a CV less than 5% between triplicate samples, we verified that our assay was both reproducible and quantitative within the element concentrations used in our study. Certified reference elements provided by the manufacturer were run in parallel to produce a standard curve. Nitrogen content was determined by the combustion method (AOAC, 2000; method 990.03) with a Leco model FP2000 (Leco Corp., St. Joseph, MI), using EDTA as a standard; and GE was determined by adiabatic bomb calorimetry with a Parr model 1261 (Parr Instrument Co., Moline, IL), using benzoic acid as a standard.
Calculations and Statistical Analyses
Apparent prececal and total tract P digestibility values were calculated using the following equation:
 | [1] |
where APD is the apparent P digestibility (prececal or total tract) expressed as a percentage; CrI is the Cr concentration of dietary intake; CrO is the Cr concentration of output (prececal digesta or feces); PO is the P concentration of output (prececal digesta or feces); and PI is the P concentration of dietary intake. All values used in this calculation were expressed as grams per kilogram of DM.
Feed intake was adjusted per feeding period to individual pig metabolic BW. Total P output [mg/(kg of BW0.75·d), either prececal or total tract] was regressed against dietary P intake [mg/(kg of BW0.75·d)] for each SBM variety using the following statistical model:
 | [2] |
where PO is expressed as mg/(kg of BW0.75·d), and TPID represents true P indigestibility. In this equation, TPID and EPL represent the slope and intercept of the simple linear regression model, respectively, whereas PO and PI represent the dependent and independent variables, respectively. True P indigestibility is simply an indirect estimate of the inefficiency with which dietary P was extracted by the pig and, therefore, TPD (expressed as %) can be calculated as:
 | [3] |
Because feed intake was adjusted to individual pig BW0.75 for each period, each pig served as its own experimental control. Therefore, a regression analysis was conducted per individual pig, resulting in a total sample size of 64 observations. Because there were 4 concentrations for each SBM, 16 observations for each TPD and EPL estimate were obtained (1 for each pig). Hence, standard errors for these regression coefficients were based on 64 individual observations.
All data were analyzed as a replicated 8 x 8 Latin square experimental design. The model for this analysis included replicate (1 df), pigs within replicate (14 df), periods within replicate (14 df), SBM variety (1 df), SBM concentration (3 df), and the interaction between SBM variety and SBM concentration (3 df). Orthogonal polynomial contrasts were used to determine the effects of SBM variety and increasing SBM concentration (and therefore, P concentration) on apparent P digestibility and total P output from growing pigs, by comparison of the response curve characteristics. Regression coefficients were compared between SBM varieties and sample collection sites (prececal and total tract samples) using confidence intervals derived from the standard errors of respective regression coefficients. Interactive effects between SBM variety and sampling site were analyzed using an orthogonal polynomial contrast of the difference between prececal and total tract measurements from individual pigs.
All data were analyzed using the GLM procedure of SAS (SAS Inst., Inc., Cary, NC). Alpha levels of 0.05 (significant) and 0.01 (highly significant) were used in all analyses, and the individual pig was considered the experimental unit. On d 1 of period 8, a pig died unexpectedly (the necropsy was inconclusive) and, therefore, missing data for period 8 were estimated for this animal. Because a Latin square experimental design was used, the estimation of missing data points was possible according to the procedures described by Montgomery (2001) and supported by Mansson and Prescott (2001).
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
The analyzed nutrient concentrations of experimental diets are presented in Table 2
. Analyzed GE content increased in sequential diets within SBM variety and ranged from 19.7 to 21.8 MJ/kg. This was due to replacement of highly digestible cornstarch with SBM, which contained more GE. Experimental diets contained increasing analyzed concentrations of CP, P, and Ca due solely to increasing concentrations of each SBM.
The average gain of barrows consuming diets formulated with conventional SBM was 1.64, 3.09, 4.00, and 4.22 kg/period for the 4 increasing SBM concentrations and 1.44, 3.06, 4.22, and 4.16 kg/period for the 4 increasing concentrations of low-phytate SBM. Linear and quadratic (P < 0.01) responses to SBM concentration were observed for weight gain regardless of SBM variety.
Dry matter intakes of barrows consuming the semi-purified SBM diets are presented in Table 3. Differences (P < 0.01) in DM digestibility were observed between prececal and total tract sampling sites. Linear decreases (P < 0.01) of apparent DM digestibility (prececal and total tract) were observed in pigs for each variety of SBM. Additionally, pigs fed low-phytate SBM exhibited a quadratic response (prececal: P < 0.05; total tract: P < 0.01) in DM digestibility.
|
for both prececal and total tract sample collection sites. Total P output determined from prececal digesta exhibited a linear relationship (P < 0.01) with increasing P concentration for each SBM variety. Additionally, linear components of the response curve were shown to differ by SBM variety (P < 0.01), thereby indicating a SBM variety x P concentration interaction. A similar linear relationship was observed for total P output in feces for conventional (linear, P < 0.01) and low-phytate SBM (linear, P < 0.01; quadratic, P < 0.05). Again, the SBM variety x P concentration interaction was highly significant (P < 0.01), suggesting differences in the response curves between SBM varieties. Increasing concentrations of dietary P, between 71.5 and 306.2 mg/(kg of BW0.75·d), increased apparent prececal P digestibility for conventional SBM (linear, P < 0.01; quadratic, P < 0.05), but not for low-phytate SBM. Conversely, P intake increased apparent total tract P digestibility of low-phytate SBM (linear and quadratic, P < 0.01) but not conventional SBM.
|
. Because of the variability, none of the endogenous P estimates was significantly different from zero. Although neither SBM variety nor sample collection site was significant for EPL, there was an interactive effect (P < 0.01). The overall EPL was 4.83 mg of P/(kg of BW0.75·d) for both SBM varieties and collection sites.
|
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
; Fan et al., 1995; Fan and Sauer, 1997). By supplying deficient dietary concentrations of P to the pig, this relationship should theoretically estimate diet-independent endogenous P excretion by extrapolation to zero dietary intake. Alternative methods to estimate EPL include feeding a P-free diet or using an isotope dilution technique. However, these methods may provide erroneous estimates due to abnormal physiological conditions of the animal or rapid recycling of isotope tracers, respectively. The regression approach balances applicability with relative experimental ease in a swine model. Typical concerns surrounding the linear regression approach include a strong dependence upon the lowest dietary concentration, which is the greatest indicator of regression curvature (Finney, 1978). Additionally, nutrient concentrations must be chosen within the linear response range to achieve accurate estimates (Fan et al., 1995; Fan and Sauer, 1997; Moughan et al., 1998). Our study was designed to reevaluate validity of the regression approach in estimating TPD and EPL in growing swine. We utilized deficient dietary P concentrations within the linear response range, similar to the approach used by Fan et al. (2001), and 2 varieties of SBM were compared. Thus, estimates were based on a relatively large sample size, and the regression approach was evaluated by comparing 2 feedstuffs within a common experiment.
The conventional and low-phytate SBM varieties used in our study had similar concentrations of CP, GE, and Ca. Reductions observed in DM digestibility as SBM replaced cornstarch, provided indirect evidence of lower SBM digestibility compared with highly digestible cornstarch in these semipurified diets, and may indicate normal gut function. The low-phytate SBM contained less total P compared with the conventional variety (5.5 vs. 6.5 g of total P/kg, respectively), and less dietary P bound to phytate. Phytate P in conventional SBM was 3.7 g/kg and accounted for 57% of total P, thereby agreeing with others (Nelson et al., 1968; Reddy et al., 1982). The low-phytate SBM contained only 1.6 g of phytate P/kg or 29% of total P; a 57% reduction in phytate P compared with conventional SBM used in our study. This is similar to the amount reported by Wilcox et al. (2000) when low-phytate mutants were first isolated. Therefore, due to similarities in nutrient concentrations between SBM varieties, it is reasonable to assume that SBM effects upon P digestibility were a direct result of differences in phytate P content.
The difference observed in TPD, and lack of difference in EPL, between conventional and low-phytate SBM lends credence to a direct phytate effect on dietary P availability in a swine model. Evaluation of low-phytate corn showed a 5-fold increase in P availability when fed to broiler chicks (Douglas et al., 2000; Li et al., 2000) and a 3-fold increase in P availability when fed to growing pigs (Veum et al., 2001), when compared with conventional corn. This improvement in P availability was shown to be a direct result of increased P digestibility. By regressing P digestibility against barley phytate P content, a 12.5% reduction in P digestibility was shown to result from each 1.0 g of phytate P/kg increase originating from barley fed to finishing pigs (Thacker et al., 2003). Sands et al. (2003) used the same low-phytate soybean mutant as used in our study (Wilcox et al., 2000), and showed 44 to 56% greater P availability for the low-phytate SBM compared with conventional SBM fed to broiler chicks. This corresponds well with the 40.7% improvement in TPD of low-phytate SBM compared with conventional SBM fed to growing pigs in our study. These studies support the hypothesis that genetic selection for low-phytate varieties can substantially reduce dietary P excretion when fed to nonruminant animals.
In our study, total P output increased with dietary P concentrations because of increasing concentrations of each SBM variety. Total P output accounted for 58 to 70% of dietary P intake for conventional SBM and 36 to 42% of P intake for low-phytate SBM when measured at the prececal collection site. However, EPL estimates in our study were quite different from those reported previously (Fan et al., 2001; Shen et al., 2002; Ajakaiye et al., 2003). An estimated EPL of 4.8 mg/(kg of BW0.75·d) was observed in our study, but this estimate was not different from zero. Fan et al. (2001) reported EPL of 78.0 and 28.0 mg of P/(kg of BW0.75·d), in prececal and total tract samples, respectively. Traylor et al. (2001) estimated prececal EPL of approximately 15.8 mg/(kg of BW0.75·d) by feeding a low protein, casein-based diet. A total tract EPL of 15.5 mg of P/(kg of BW0.75·d) was obtained by summarizing 66 P balances and using regression analysis (Rodehutscord et al., 1998). In all studies, suboptimal concentrations of dietary P were fed, and Rodehutscord et al. (1998) concluded that with deficient P intake, utilization of digestible P was almost complete. Rodehutscord et al. (1998) also provided evidence that 1) a strong relationship exists between P excretion and P intake, and 2) fecal EPL is correlated with BW, but not DMI. For these reasons, EPL estimates herein have been reported relative to metabolic BW of the pig.
To highlight the variability observed in EPL, a summary of estimates reported in the literature can be found in Table 6. This summary provides mean BW of pigs studied, dietary P concentrations utilized, and EPL estimates obtained by various methodologies. Additionally, this compilation of estimates shows that recent estimates are relatively low compared with older research. Using the whole-body composition technique, Pettey et al. (2004) estimated EPL of 9.3, 7.2, and 7.2 mg/(kg of BW0.75·d) for pigs weighing 27, 59, and 98 kg, respectively. Petersen and Stein (2004) reported a fecal EPL of 10.8 mg of P/(kg of BW0.75·d) when feeding a purified cornstarch diet devoid of P. Therefore, accumulation of recent diet-independent EPL estimates suggests that inevitable loss of P is rather small; likely less than 20 mg of P/(kg of BW0.75·d). Excretion of P above this amount may be attributed to diet-dependent sources. We used the same regression technique to estimate endogenous losses as Fan et al. (2001), Shen et al. (2002), and Ajakaiye et al. (2003), and found no explanation for the discrepancy in EPL estimates. Because we obtained TPD estimates for conventional SBM that were similar to the results of Fan et al. (2001), it seems that the regression approach is repeatable between laboratories for TPD estimates.
|
In our study, P and Ca were supplied at increasing concentrations below the animal’s requirement. The average Ca:total P ratio for conventional and low-phytate SBM diets was 0.64:1 and 0.88:1, respectively, due to both Ca and P being supplied solely by SBM. Al-Masri (1995) showed that EPL was negatively correlated with dietary Ca concentrations and wide Ca:total P ratios in broiler chicks. Data on the effect of Ca:total P ratios less than 1:1 with deficient P and Ca concentration are limited, but Fan et al. (2001) used Ca:total P ratios and Ca concentrations similar to those used in our experiment. High Ca:total P ratios (i.e., 1.5:1 or above) have been shown to reduce P utilization in low-P corn-SBM diets for growing-finishing pigs (Liu et al., 1998, 2000). It is unknown whether deficient dietary P and Ca concentrations or a Ca:total P ratio less than 1:1 affected P utilization or EPL in our study. Because we did not collect urine, it is unknown whether significant amounts of P were excreted via this route. However, urinary P excretion at the dietary P concentrations used in our study (below requirements) would likely be negligible (Rodehutscord et al., 1998). Rodehutscord et al. (1999) showed that urinary P excretion was 12 to 25 mg/d when P-depleted pigs were fed increasing intakes of P (2.2 to 8.6 g/d) over a 60-d period. The P intake in our study was 0.9 to 3.7 g/d.
If provided at the recommended total P requirement of 6.0 g/d for the 10- to 20-kg pig (NRC, 1998), the EPL estimate from our study [4.8 mg/(kg of BW0.75·d)] would account for only a small proportion of required P intake. Clearly, estimates of EPL were variable in our study, possibly resulting from variation in apparent P digestibility between pigs (Rodehutscord et al., 1997, 1998). The degradation of dietary phytate by swine hindgut microflora may also explain variation in P utilization within and between SBM varieties. In contrast to Fan et al. (2001), we observed no difference between prececal and total tract EPL estimates when feeding either conventional or low-phytate SBM. Therefore, our data suggest that there is limited absorptive capacity for P in the cecum and colon of growing swine.
Our results agree with Sands et al. (2003) that low-phytate SBM has greater available P than conventional SBM. Therefore, we conclude that increased TPD of low-phytate SBM fed to growing pigs is the direct result of the 57% reduction in analyzed phytate P content of low-phytate SBM vs. conventional SBM.
|
IMPLICATIONS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
|
Footnotes |
|---|
2 Present address: Department of Animal Sciences, University of Illinois, Urbana 61801
3 Corresponding author: ladeola{at}purdue.edu
Received for publication June 2, 2005. Accepted for publication November 14, 2005.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. S. Akinmusire and O. Adeola True digestibility of phosphorus in canola and soybean meals for growing pigs: Influence of microbial phytase J Anim Sci, March 1, 2009; 87(3): 977 - 983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Schulin-Zeuthen, E. Kebreab, W. J. J. Gerrits, S. Lopez, M. Z. Fan, R. S. Dias, and J. France Meta-analysis of phosphorus balance data from growing pigs J Anim Sci, August 1, 2007; 85(8): 1953 - 1961. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. I. Petersen and H. H. Stein Novel procedure for estimating endogenous losses and measurement of apparent and true digestibility of phosphorus by growing pigs J Anim Sci, August 1, 2006; 84(8): 2126 - 2132. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
