|
|
||||||||
ANIMAL NUTRITION |

* Department of Animal Sciences, Purdue University, West Lafayette, IN 47907-2054; and
and
Danisco Animal Nutrition, Marlborough, SN8 1XN, UK
| Abstract |
|---|
|
|
|---|
Key Words: broiler equivalency Escherichia coli monosodium phosphate phytase pig
| INTRODUCTION |
|---|
|
|
|---|
The current study was conducted to evaluate the efficacy of an E. coli-derived phytase and determine its equivalency relative to P from monosodium phosphate (MSP) using growth performance and nutrient digestion in broiler chickens and young pigs.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Efficacy and Equivalency in Broilers (Exp. 1)
One thousand two hundred 1-d-old male broiler chicks were weighed individually, sorted based on BW, and assigned to 8 diets, such that the average BW across dietary treatments was similar. There were 6 replicate pens with 25 birds per pen. A schedule of 23:1 h of light to dark was maintained. Pen temperatures from d 0 to 7, 8 to 14, and 15 to 42 were 35, 32, and 27°C, respectively. Pens with wood shavings were 1.5 x 2.4 m, providing 0.14 m2/bird.
The chicks were fed starter diets for 3 wk. The starter diets were based on a basal diet (BD) formulated to contain 3,200 kcal of ME, 224 g of CP, 7.7 g of Ca, and 2.4 g of nonphytate P (nPP) per kilogram and adequate in all other nutrients, according to the recommendations of the NRC (1994
; Table 1
). The BD also contained 2.5 g of Cr2O3 per kg as an indigestible marker. The starter diets consisted of the BD; the BD plus 0.6, 1.2, or 1.8 g of inorganic P (iP) from MSP; and the BD plus 250, 500, 750, or 1,000 FTU per kg of an E. coli-derived phytase. The phytase used in the present research was an E. coli-derived phytase expressed in Schizosaccaromyces pombe (Phyzyme XP, Danisco Animal Nutrition, Marlborough, UK). All diets were fed in a mash form, and the phytase was added as a powder.
|
On d 21, 5 chicks closest to the average weight for each pen were selected as a group and assigned to battery cages for excreta collection as described by Adedokun et al. (2004)
. A schedule of 23:1 h of light to dark was maintained, and battery temperatures were 27°C. From d 21 to 23, excreta samples were collected from each cage. On d 23, chicks were euthanized via CO2 inhalation. Digesta samples were collected from the ileum, and the tibia was taken from each bird for analysis. Also on d 21, floor pens were normalized to 18 birds/pen to account for losses in some pens.
Efficacy and Equivalency in Young Pigs (Exp. 2)
Pigs were weaned at approximately 21 d and fed a diet containing 220 g of CP and 11.5 g of Lys per kg of diet for 14 d before the pigs were selected for use in the experiment. Eight dietary treatments consisted of a positive control (PC) containing 7.1 g of Ca, 2.4 g of nPP per kg with no supplemental phytase and adequate in all nutrients (NRC, 1998
; Table 1
), a negative control (NC) containing no supplemental phytase or iP with 5.3 g of Ca, 0.8 g of nPP per kg, and a Ca:P ratio similar to that of the PC; the NC plus 0.4, 0.8, or 1.2 g of P from MSP per kg; and the NC plus an E. coli-derived phytase at 500, 750, or 1,000 FTU per kg. All diets contained 3.0 g of Cr2O3 per kg as an indigestible marker.
Forty-eight pigs weighing approximately 10 kg were sorted based on BW and sex and housed individually in stainless-steel pens. Pigs were assigned to 8 dietary treatments with 6 pigs per treatment. The pens (0.76 x 0.89 m), each equipped with a nipple waterer, stainless-steel feeder, and plastic-coated expanded metal floor, were located in an environmentally regulated building maintained at 23 ± 2°C, with a 12-h light to dark cycle.
Pigs were weighed and feed intake was recorded once a week. Fresh fecal grab samples were collected on d 22 to 28, pooled by pig, and stored at 18°C until analysis. On d 28, pigs were euthanized, and the left third and fourth metacarpals were excised for determination of bone ash.
Chemical Analysis
Excreta and fecal samples were dried at 55°C in a forced-draft oven, and ileal digesta samples were freeze-dried (Freezemobile 12SL, The Virtis Co. Inc., Gardiner, NY) for 5 d. Excreta, fecal, ileal, and feed samples were ground through a 1-mm screen. Samples were then used to determine DM content by oven drying at 105°C for 24 h. Nitrogen content of the diets was determined by the combustion method (model FP2000, Leco Corp., St. Joseph, MI) according to the AOAC (2000)
, and GE was determined by adiabatic bomb calorimetry (model 1261, Parr Instrument Co., Moline, IL).
Samples from broilers were analyzed for AA, Ca, Cr, and P (Missouri Experiment Station Chemical Laboratory, University of Missouri, Columbia). Samples for AA analysis were prepared by a 24-h hydrolysis in 6 N HCl at 110°C under an N atmosphere. For Met and Cys analysis, samples were oxidized using performic acid before acid hydrolysis. Samples for Trp analysis were hydrolyzed using barium hydroxide. Amino acids in the hydrolysate were determined by HPLC after postcolumn derivatization (AOAC, 2000
). Amino acid concentrations were not corrected for incomplete recovery resulting from hydrolysis. Calcium, Cr, and P concentrations of samples were determined by the inductively coupled plasma, atomic emission spectroscopy method (AOAC, 2000
) after wet-ash digestion with nitric and perchloric acids.
Fecal and feed samples from pigs were prepared using wet-ash digestion (AOAC, 2000
), and Cr was determined by measuring absorption at 540 nm (Spectronic 21D, Milton Roy Company, Rochester, NY). Calcium in wet-ashed samples was determined by an atomic absorption spectrometer (AAnalyst 300, Perkin Elmer, Norwalk, CT). Phosphorus concentration was determined using a colorimetric assay. Acid molybdate and Fiskes SubbaRow reducer solution were added to wet-ash samples to form a phosphomolybdenum complex. Color intensity was proportional to P concentration and was determined with a spectrophotometer using absorbance at 620 nm (AOAC, 2000
; SpectraCount, Model # AS1000, Packard, Meriden, CT). Phytase activity was determined by the method of Engelen et al. (1994)
. One FTU is defined as the quantity of enzyme required to hydrolyze 1 µmol of iP from an excess of 15 µM sodium phytate per min at pH 5.5 and 37°C (International Union of Biochemistry, 1979
). Left metacarpal bones from pigs and broiler tibias were excised at slaughter, defleshed by blunt dissection, lipid-extracted with ether, and ashed for 16 h at 600°C in a muffle furnace to determine bone ash.
Statistical Analysis
All data were evaluated for violation of the basic assumptions for ANOVA and transformed by the Box-Cox method (Montgomery, 2001
) when necessary. Growth performance, digestibility, retention, and bone ash data were analyzed by the GLM procedure (SAS Inst. Inc., Cary, NC) appropriate for a randomized complete block design. Untransformed means are reported, but for subsequent analysis, transformed values were substituted where necessary, and means were compared by orthogonal polynomial contrasts. Linear and nonlinear regression analyses were conducted on several response criteria using the JMP statistical package from SAS. The linear regression model used was
![]() |
where a is the y intercept, b is the slope of the line, and X is level of supplemental iP from MSP or FTU from E. coli-derived phytase per kg. The nonlinear regression model used was the mechanistic growth model:
![]() |
where
1 is the y maximum,
2 controls the y intercept,
3 controls the degree of curvature, and X is level of supplemental iP from MSP or FTU from E. coli-derived phytase per kg. Dependent variables (response criteria) were regressed against independent variables (calculated iP from supplemental MSP or phytase intake from supplemental E. coli-derived phytase). Phosphorus equivalency of phytase was determined through solution of the equations for the independent variable at an equivalent level of the dependent variable.
The second-order Akaikes Information Criterion (AICc) was used to determine whether the linear or nonlinear regression was a better fit to the data (Motulsky and Christopoulos, 2003
). The AICc is calculated for each line using the equation
![]() |
where N is the number of data points, Ln is the natural log, K is the number of parameters fitted by the regression equation plus 1, and SS is the error sums of squares. The probability that the nonlinear regression equation is correct was calculated using the following equation:
![]() |
where
equals the AICc of the nonlinear equation minus the AICc of the linear equation. The sum of the probabilities for the 2 equations explaining the same data is 1, with a probability
0.5 indicating a nonlinear curve is appropriate, probabilities
0.5 indicating a linear curve is more appropriate, and probabilities around 0.5 considered ambiguous. If the data for 1 supplement were better fit by a linear equation and the data for the other supplement were better fit by the nonlinear equation, the supplement with the higher probability determined the set of equations to be used.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
|
In Exp. 2, the phytase activity of the diets formulated to contain 500, 750, or 1,000 FTU per kg was determined to be 586, 832, or 1,226 FTU per kg, respectively. Pigs fed the NC diet had lower average daily gain, final BW, and percent bone ash than pigs fed the PC diet (P < 0.05; Table 7
). Increasing levels of MSP resulted in linear increases in ADG, final BW, and percent bone ash (P < 0.05). Increasing supplemental E. coli-derived phytase increased ADG (linear and quadratic, P < 0.05) and percent bone ash (linear, P < 0.05).
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Traditional industry diets use dicalcium phosphate as the source of supplemental iP. However, NRC (1994
However, NRC (1998)
states that dicalcium phosphate is 95 to 100 percent bioavailable, with bioavailability generally expressed as a percentage of MSP or monocalcium phosphate. Johnston et al. (2004)
, Liu et al. (2000)
, and Ballam et al. (1984)
reported that the efficacy of phytase can be influenced by the ratio of Ca to P, with the efficacy decreasing dramatically when Ca concentration is 2 or more times that of P. In an attempt to minimize the possibility of confounding effects, we decided to use MSP as our iP reference.
The weight gain response of birds to phytase supplementation indicated that the efficacy of this enzyme could peak or plateau around 500 FTU per kg because of the diminishing response to phytase supplementation. This is supported by the fact that the nonlinear equation was 6.5-fold (86.73/13.27%) better fit than the linear equation (Motulsky and Christopoulos, 2003
). Interestingly, Onyango et al. (2005)
did not see a plateau; instead they reported 17 and 20% increases in gain per bird from d 8 to 22 by supplementing a low-P diet with 500 and 1,000 FTU per kg of an evolved-E. coli phytase, respectively. Dilger et al. (2004)
reported 6 and 19% improvements in gain from d 8 to 22 with 500 and 1,000 FTU of the same E. coli-derived phytase used in the current study per kg, respectively, and 9, 14, and 14% increases in gain from d 0 to 42 in male broiler chicks with 500, 750, and 1,000 FTU of Phyzyme XP per kg, respectively. These results, in conjunction with the results of Onyango et al. (2005)
, indicate that the plateau may not be observed within the range of 0 to 1,000 FTU per kg in broiler chicks fed an adequate P diet from d 0 to 7 and then the phytase-supplemented diet from d 8 to 22. With such a design, chicks build up a sufficient P reserve, thus possibly masking the plateau we have observed at higher supplementation levels.
Addition of phytase to broiler diets increased ileal digestibility of 4 essential and most of the nonessential AA on d 21. Adeola and Sands (2003)
documented some instances in the literature where small increases in digestibility of AA were observed. However, those observed increases rarely translated into improved growth performance after accounting for the effect of improved phytate P availability. Adeola and Sands (2003)
also pointed to a larger body of reports in which no response was observed in AA utilization to phytase supplementation.
Effects on apparent retention in boilers were far less pronounced at d 21. Increasing levels of MSP negatively affected P retention. Addition of phytase improved retention of P but decreased retention of DM and energy.
Denbow et al. (1995)
, Harper et al. (1997)
, and Ravindran et al. (2001)
used r2 as the criterion for selecting either linear or nonlinear regression model to describe the changes in certain response criteria to graded levels of phytase and supplemental iP. Neter et al. (1996)
, however, indicated that r2 is not a meaningful statistic for nonlinear regression because the product of the error sum of squares and the regression sum of squares is not necessarily equal to the total sum of squares. Motulsky and Christopoulos (2003)
indicated that the 2 best methods for comparing 2 regression equations for goodness of fit of the same data are an F-test if the 2 equations are nested or AICc if they are not. Because it is not possible to get the linear equation by setting some of the parameters of the mechanistic growth model (Y =
1 x (1
2 x e(
3 x X))), , to zero or one, the AICc is the only valid means of comparison in this situation.
The r2 values of linear models generated from the poultry growth performance data were lower than those observed in the literature (Denbow et al., 1995
; Yi et al., 1996a
; Adedokun et al., 2004
). The differences may be attributed to duration of study and age of birds. Denbow et al. (1995)
and Yi et al. (1996a)
conducted 21-d trials, whereas our trial was 42 d in duration. Adedokun et al. (2004)
used only linear regression equations and obtained high r2 values using the same E. coli-derived phytase. However, the duration of their study was 14 d after allowing birds ad libitum access to a standard P-adequate starter diet for 7 d. The higher r2 values of Adedokun et al. (2004)
could be the result of P stores accumulated during the first 7 d, which may have decreased variability by modifying the response to lower iP in the diet. They could also be the result of a shorter observational period because we had 4 more weeks for variance to increase as the birds grew. The equivalent of between 0.92 and 1.82 g of P from MSP would be released by 1,000 FTU of E. coli-derived phytase per kilogram of diet in broilers depending on the performance criteria used. The most reliable predictor of equivalency in this study, based on a combination of highest r2 value and probability from AICc, was gain per bird with 1,000 FTU per kg being equivalent to 0.92 g of iP from MSP per kg. This is slightly lower than 1.01, or 1.03 g of P per kg for BW gain and the average of all response criteria, respectively, reported by Adedokun et al. (2004)
.
The weanling pigs responded to supplemental MSP or E. coli-derived phytase with linear improvements in ADG, percent bone ash, and BW at the end of 4 wk of study. Using a different E. coli-derived phytase for 28 d in weanling pigs, Adeola et al. (2004)
observed 17 and 10% increases in ADG and final BW, respectively, over the NC with 1,000 FTU per kg, compared with improvements of 19 and 8% observed for those response criteria, respectively, in the current study. However, the greatest difference in ADG, percent bone ash, or final BW between the NC and phytase supplementation was observed with the NC + 750 FTU per kg of diet, indicating the diminishing response to phytase supplementation in pigs as well. Similar responses were observed in the nutrient digestibility. Phosphorus digestibility increased by 0.22, 0.13, and 0.02 as phytase supplementation increased from 0 to 500, 500 to 750, and 750 to 1,000 FTU per kg, respectively.
The r2 values for the performance prediction equations generated from the weanling pig growth performance data were lower than those reported by Yi et al. (1996b)
and Harper et al. (1997)
. Those researchers developed regressions using treatment means rather than the individual observations, which may have resulted in elevated r2 values. Harper et al. (1997)
reported that 500 units of fungal-derived phytase was equivalent to the addition of between 0.87 and 0.96 g of iP from dicalciummonocalcium phosphate supplements, whereas Yi et al. (1996b)
reported that it was equal to the addition of 1.28 g of iP from defluorinated phosphate. Five hundred FTU of E. coli-derived phytase per kg used in the current study was found to be equivalent to the addition of 0.49 and 1.00 g of iP from MSP per kg of diet based on ADG and bone ash, respectively. The differences between our results and those of Yi et al. (1996b)
and Harper et al. (1997)
could be greater than reported because of the use of different sources of iP as standards. Phosphorus availability is similar for monocalcium phosphate and MSP (NRC, 1998
), but de-fluorinated phosphate is only 80% available compared with MSP (Soares, 1995
). Kemme et al. (1997)
indicated that the efficacy of phytase in releasing digestible P is influenced by the physiological status of pigs with grow-finishing pigs ranking above starter pigs. Phytase has been shown to be more efficacious in reducing water-soluble P excretion in grower pigs than starter pigs (J. A. Jendza and O. Adeola, unpublished data). Starter pigs may lack the ability to adequately capitalize on phytase supplementation due to a lack of gut maturity or insufficient retention time.
In summary, the E. coli-derived phytase used in the current study is capable of partially alleviating the deleterious effect of feeding a P-deficient corn-soybean meal-based diet on growth performance in broiler chicks and weanling pigs, as well as influencing the digestibility of several essential AA in broilers. When supplementing P-deficient diets with 500 FTU per kg, the E. coli-derived phytase is equivalent to the addition of between 0.72 and 1.19 g in broiler diets and between 0.49 and 1.00 g of iP from MSP per kg in weanling pig diet.
| Footnotes |
|---|
2 Corresponding author: ladeola{at}purdue.edu
Received for publication April 5, 2006. Accepted for publication July 18, 2006.
| LITERATURE CITED |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. L. Veum and M. R. Ellersieck Effect of low doses of Aspergillus niger phytase on growth performance, bone strength, and nutrient absorption and excretion by growing and finishing swine fed corn-soybean meal diets deficient in available phosphorus and calcium J Anim Sci, April 1, 2008; 86(4): 858 - 870. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |