Supplementing corn or corn–barley diets with an E. coli derived phytase decreases total and soluble P output by weanling and growing pigs A. D. Beaulieu1, M. R. Bedford2, and J. F. Patience1 1Prairie

Swine Centre, Inc. PO. Box 21057, 2105 8th St. E, Saskatoon, Saskatchewan, Canada S7H 5N9 (e-mail: [email protected]); 2Syngenta Animal Nutrition, Wiltshire, UK (e-mail: [email protected]). Received 21 November 2006, accepted 3 April 2007.

Beaulieu, A. D., Bedford, M. R. and Patience, J. F. 2007. Supplementing corn or corn–barley diets with an E. coli derived phytase decreases total and soluble P output by weanling and growing pigs. Can. J. Anim. Sci. 87: 353–364. The efficacy of an E. coli derived phytase on phosphorus (P) digestibility and excretion, on the form of the P excreted, and the optimal dietary calcium (Ca):P ratio was examined. In exp. 1, 63 barrows (40.4 ± 1.9 kg) were assigned to receive one of 21 treatments arranged as a 3 × 7 factorial. Treatments consisted of three Ca levels (0.50, 0.60 and 0.70%) and seven phytase treatments [0, 250, 500, 1000 and 2000 FTU kg–1 of an E. coli-derived phytase and 500 FTU kg 1 of an A. niger phytase added to a P deficient (0.37% P) diet or 0 FTU phytase kg–1 added to a P adequate (0.53% P) diet]. In exp. 2, 144 pigs (6.52 ± 0.75 kg), received a P adequate (0.60% P) diet or a P deficient (0.44% P) diet supplemented with 0, 250, 500, 1000 or 2000 FTU phytase kg–1 for a 28-d trial. A subset of 36 barrows was then fed the same diets in a balance trial. In exp. 3, 36 barrows (7.1 ± 0.75 kg) were assigned to one of six treatments arranged as a 2 × 3 factorial (0 or 500 FTU of phytase kg–1; 1.0, 1.6 or 2.2 Ca:P ratio). In exp. 1, P digestibility improved from 21 to 54% with increasing phytase (quadratic; P < 0.05). Supplementing the diet with 500 FTU phytase kg–1 decreased the output of total and soluble P by 25% in exp. 2 and to a similar extent in exp. 3 at the lowest Ca:P ratio (P < 0.05). The effect of phytase on total P digestibility was mitigated as the dietary Ca:P ratio increased in exp. 3. Supplementation of swine diets with an E. coli derived phytase decreases output of total and soluble forms of P, but this effect is reduced at high dietary Ca:P ratios. Key words: Swine, E. coli phytase, phosphorus, soluble phosphorus Beaulieu, A. D., Bedford, M. R. et Patience, J. F. 2007. L’enrichissement d’une ration de maïs ou de maïs et d’orge avec une phytase dérivée de E. coli réduit l’excrétion du P total et du p soluble chez les porcelets sevrés et les porcs en croissance. Can. J. Anim. Sci. 87: 353–364. Les auteurs se sont penchés sur l’efficacité avec laquelle une phytase dérivée de E. coli agissait sur la digestibilité et sur l’excrétion du phosphore (P). Ils ont aussi déterminé le ratio optimal entre le calcium (Ca) et le phosphore dans les aliments. Lors d’une première expérience, ils ont réparti 63 castrats (40,4 ± 1,9 kg) entre 21 traitements dans une disposition factorielle 3 × 7. Les traitements consistaient en 3 concentrations de Ca (0,50, 0,60 et 0,70 %) et 7 de phytase (0, 250, 500, 1 000 et 2000 FTU par kg de phytase dérivée de E. coli et 500 FTU par kg d’une phytase de A. niger dans une ration carencée en P [0,37 %] ou 0 FTU de phytase par kg dans une ration à concentration suffisante de P [0,53%]). Dans le cadre d’une deuxième expérience, les auteurs ont donné à 144 porcs (6,52 ± 0,75 kg) une ration contenant suffisamment de P (0,60 %) ou carencée en P (0,44 %) et enrichie avec 0, 250, 500, 1 000 ou 2 000 FTU de phytase par kg pendant 28 jours. Un sous-échantillon de 36 castrats a ensuite reçu la même ration dans le cadre d’un essai sur le bilan alimentaire. Enfin, dans une troisième expérience, 36 castrats (7,1 ± 0,75 kg) ont été répartis entre six traitements dans une disposition factorielle 2 × 3 (0 ou 500 FTU de phytase par kg; ratio Ca:P de 1,0, 1,6 ou 2,2). Dans la première expérience, la digestibilité du P est passée de 21 % à 54 % avec le relèvement de la dose de phytase (quadratique; P < 0,05). Ajouter 500 FTU de phytase par kg à la ration a diminué l’excrétion de P total et de P soluble de 25 % dans la deuxième expérience et presque autant dans la troisième, au plus faible ratio Ca:P (P < 0,05). Dans cette dernière expérience, la hausse du ratio Ca:P atténue l’incidence de la phytase sur la digestibilité du P total. Enrichir l’alimentation des porcs avec une phytase dérivée de E. coli diminue l’excrétion du P total et du P soluble, mais dans une moindre mesure lorsque le ratio Ca:P est élevé. Mots clés: Porcs, phytase de E. coli, phosphore, phosphore soluble

however, exhibit greater sorption to soils and thus may pose less labile than other forms of P (Leytem et al. 2002). Phytases (myo-inositol hexakisphosphate phosphohydrolases), which occur to some extent in feedstuffs and are produced by various micro-organisms, hydrolyze phytic acid to

Fifty to seventy percent of the P in grains and oilseeds commonly used in swine diets is organically bound as salts of phytic acid (Eeckhout and De Paepe 1994). Due to a lack of sufficient endogenous phytase activity to degrade the phytate, the majority of the feed P is not utilized by the pig, but is excreted into the manure and the P requirement of the pig has been met with the addition of inorganic P (National Research Council 1998). Manure P has been identified as a significant source of soil P enrichment in areas of high livestock density (Daumer et al. 2004). Organic P compounds,

Abbreviations: ADFI, average daily feed intake; ADG, average daily gain; aP, available phosphorus; BW, body weight; FCE, feed efficiency (gain to feed); tP, total phosphorus 353

354

CANADIAN JOURNAL OF ANIMAL SCIENCE

inositol and orthophosphate (Eeckhout and De Paepe 1994). It has been well documented that supplementing swine diets with exogenous phytase improves the utilization of phytate bound P (Lei et al. 1993a, b); however, it has also been suggested that the release of orthophosphates by phytase may increase the soluble P fraction of swine manure (Daumer et al. 2004). This fraction of P in swine manure is more prone to runoff, which can potentially lead to increased eutrophication of groundwater (DeLaune et al. 2004). Commercially available phytases have been primarily 3or 6-phytases, referring to the location on the molecule where dephosphorylation is initiated (Greiner et al. 2002). A 6-phytase, isolated from E. coli exhibited a pH optima of 2.5 (Rodriguez et al. 1999) and improved resistance to the porcine proteases, pepsin and pancreatin, relative to phytases derived from Aspergillus sp. (Simon and Igbasan 2002). When added to corn and soybean meal-based diets of weanling (Stahl et al. 2000; Adeola et al. 2004), growing, or finishing pigs (Jendza et al. 2005), the E. coli derived phytase effectively released phytate P as evidenced by improved performance, increased bone ash (Stahl et al 2000; Jendza et al. 2005), and increased P digestibility and retention (Adeola et al. 2004). The efficacy of A. niger phytase in corn and soybeanmeal-based diets was mitigated at high dietary Ca:P ratios (Lei et al. 1994; Liu et al. 1998). Similarly, the beneficial effects of a P. lycii phytase on reducing faecal P excretion, observed when the Ca:P ratio in the diet was 1.15:1, was no longer evident when added to a diet with a ratio of 1.85:1 (Brady et al. 2002). The primary objective of the following studies was to determine the efficacy of an E.coli derived phytase when added to corn and soybean meal, or barley, corn and soybean-meal-based diets on P digestibility and excretion. Additionally, we wanted to determine (1) the Ca:P dietary ratio that is optimal with the use of an E. coli derived phytase and (2) whether the form (solubility) of the P in the manure was altered with the use of phytase. MATERIALS AND METHODS General The protocol for the following experiments was approved by the University of Saskatchewan, University Committee on Animal Care and Supply, and followed guidelines established by the Canadian Council on Animal Care (1993). Euthanasia was accomplished by captive bolt stunning followed by exsanguination. All animals were housed in rooms with automatic control of ventilation and temperature set to maintain a thermoneutral environment. Lights were on from 0700 to 1900. Except where noted, all diets were formulated to meet or exceed nutrient requirements for the stated category of pig (National Research Council 1998). Representative diet samples were collected weekly, composited by treatment and submitted to a commercial lab (Norwest Labs, Lethbridge, AB) for analysis of N [Kjeldahl (Association of Official Analytical Chemists) AOAC 988.05], crude fibre (Ankom; AOAC 962.09), and K and Na using inductively coupled plasma atomic emission spec-

trophotometry (ICP; AOAC 985.01). These data were used to verify agreement of diet specifications with formulated values and thus ensure that requirements of the stated age of pig were met, and are not reported. Diet supplementation of the E. coli derived phytase (Quantum; Syngenta Animal Nutrition, Research Triangle Park, NC) was intended to provide approximately the same amount of available P (aP) as the positive control based on actual activity previously determined for that specific lot by a commercial lab according to the method of Engelen et al. (2001). One phytase unit (FTU) is defined as the amount of enzyme required to release 1 µmol of inorganic P per minute from sodium phytate at pH 5.5 and 37ºC. In all experiments, an initial pool of pigs was selected based on weight per day of age. Pigs were then assigned to treatment based on body weight (BW). Data were analyzed using the MIXED procedure of SAS (SAS Institute, Inc., Cary NC). The Kenward-Roger option was used to estimate denominator degrees of freedom. Unless indicated otherwise, data are presented as least-squares means (LSMeans). Significance was declared at P < 0.05 and a trend is discussed if P > 0.05 but < 0.10. Specific models are described for each experiment. Experiment 1 Animals and Diets Individually housed, castrated male pigs (n = 63) weighing 40.3 ± 1.9 kg (mean ± SD) were assigned by BW in three blocks to one of 21 treatments for a 28-d experiment. Individual pig weights were recorded at the start of the trial and every 7 d thereafter. All feed added to the feeders was recorded and feeder weigh backs recorded weekly for the calculation of feed disappearance used to estimate feed intake. Grab faecal samples were collected on days 27 and 28. The two daily samples were combined within experimental period by pen, lyophilized, and analyzed for acidinsoluble ash and total phosphorus (tP) using methods described below for diets. The corn and soybean meal based diets are shown in Table 1. Treatments were arranged as a 3 × 7 factorial, and consisted of 3 Ca levels (0.50, 0.60, 0.70%) and a series of seven phytase treatments which consisted of the negative control (no supplemental P, no added phytase), a positive control (no added phytase) and four levels (250, 500, 1000 and 2000 FTU kg–1) of an experimental E.coli-derived phytase (2900 FTU g–1, Quantum; Syngenta Animal Nutrition, Research Triangle Park, NC) or 500 FTU kg–1 of a commercially available A. niger phytase (min 1000 FTU g–1, Natuphos® 1000 G, BASF, Georgetown, ON) added to the negative control. The negative control, formulated with supplemental inorganic P, contained 0.41% tP and an estimated 0.11% aP. The positive control contained supplemental dicalcium phosphate and was formulated to contain 0.50% tP and 0.23% aP, meeting the requirements for pigs of this age (National Research Council 1998). Celite (0.4%; Celite Corp. CA) was added to the diets as a source of acid-insoluble ash that was analyzed gravimetrically by a commercial lab (Norwest Labs, Lethbridge, AB) using AOCS Method Ba–68 (American Oil Chemists Association 1993). All diets

BEAULIEU ET AL. — E. COLI DERIVED PHYTASE DECREASES TOTAL AND SOLUBLE P OUTPUT

355

Table 1. Ingredient and nutrient composition of experimental diets (as fed basis) Experiment 1zy Negative control Ingredient (%) Corn Barley Soybean meal Limestone Spray-dried whey Dicalcium phosphate Salt PSC mineral premixv PSC vitamin premixu l-lysine HCl l-threonine dl-methionine Choline chloride l-tryptophan Canola oil Celite Chromic oxide LS20s Nutrients DE (Mcal kg–1) (formulated) dLys g Mcal–1 DE (formulated) Total P (%) Calcium (%)

70.08 – 27.27 0.64 – 0.00 0.40 0.50 0.50 0.17 0.04 – – – – 0.40 – – 3.20 2.60 0.37r 0.52q

Experiment 2x

Positive control

Negative control

Positive control

69.80 – 27.16 0.36 – 0.68 0.40 0.50 0.50 0.17 0.04 – – – – 0.40 – –

40.00 29.80 27.00 1.10 – 0.25 0.33 0.50 0.50 0.30 0.09 0.05 0.08 0.02 – – (0.40)t –

39.47 29.80 27.00 0.78 – 1.10 0.33 0.50 0.50 0.30 0.09 0.05 0.08 0.02 – – (0.40)t –

3.53 3.20 0.44p 0.70p

3.53 3.20 0.60p 0.70p

3.20 2.60 0.53r 0.51q

Experiment 3 Ca:tP ratiow 1.0

1.6

2.2

54.95 15.00 20.80 0.40 5.00 0.60 0.40 0.50 0.50 0.29 0.11 0.04 0.08 0.02 1.00 – 0.40 0.10

54.17 15.00 20.80 1.18 5.00 0.60 0.40 0.50 0.50 0.29 0.11 0.04 0.08 0.02 1.00 – 0.40 0.10

53.39 15.00 20.80 1.97 5.00 0.60 0.40 0.50 0.50 0.29 0.11 0.04 0.08 0.02 1.00 – 0.40 0.10

3.46 2.70 0.52o 0.56o

3.46 2.70 0.52o 0.86o

3.46 2.70 0.52o 1.17o

zDiets shown are for the 0.50% Ca series. The Ca concentration in the 0.60% and 0.70% Ca series was increased with the addition of limestone added at the expense of corn. yFormulated to be either limiting (negative control) or adequate (positive control) in aP for 40 kg pigs. Either 250, 500, 1000 or 2000 FTU kg–1 of an experimental E. coli derived phytase (Quantum phytase; Syngenta Animal Nutrition, Research Triangle Park, NC) or 500 FTU kg–1 of a commercial A. niger phytase (Natuphos® 1000 G, BASF, Georgetown, ON) was added to the negative control. xFormulated to be either limiting (negative control) or adequate (positive control) in aP for 10 kg pigs. Either 250, 500, 1000 or 2000 FTU kg–1 of an experimental E. coli derived phytase enzyme was added to the negative control. wBased on analyzed values; 500 FTU kg–1 an experimental E. coli derived phytase enzyme was added to each Ca:tP ratio. vProvided (per kg of diet); Zn, 100 mg as zinc sulphate; Fe, 80 mg as ferrous sulphate; Cu, 50 mg as copper sulphate; Mn 25 mg as manganous sulphate; I, 0.50 mg as calcium iodate; Se, 0.10 mg as soldium selenite. uProvided (per kg of diet), vitamin A, 8,250 IU; vitamin D 825 IU; vitamin E, 40 IU; niacin, 35 mg; D-pantothenic acid, 15 mg; 5 mg; menadione, 4 mg; 3 folacin, 2 mg; thiamine, 1 mg; D-biotin, 0.2 mg; vitamin B12, 25 µg. tAdded to the diet for the balance study only. sContains lincomycin at 22 g kg–1 and spectinomycin at 22 g kg–1 (BioAgrimix, Mitchell, ON). rMean of analyzed values. Negative control, n = 18 (SD = 0.017); positive control, n = 3 (actual values; 0.53%, 0.52%, 0.54%). qValues shown are analyzed values of the negative and positive control for the 0.50% Ca series. Corresponding values for the negative and positive control for the 0.60% series; 0.64% and 0.64%, and for the 0.70% series; 0.85% and 0.76%, respectively. pFormulated. Analyzed values in Table 3. oMean of two diets, negative control and the negative control plus phytase.

were analyzed for tP, and the positive and negative control diet from each Ca level were analyzed for Ca by ICP as described for K and Na in the general methods section. Statistical Analysis Production data were analyzed using a randomized complete block design with week as a repeated measure and the individual pig as the experimental unit. The model contained the fixed effects of enzyme amount, dietary Ca and the interaction of enzyme by dietary Ca. Initial weight was used as a covariate. The variance-covariance matrix was chosen for each variable based on fit statistics within SAS. The Pdiff option within SAS was used to compare: (1) the response to phytase source (E. coli vs. A. niger; 500 FTU kg–1 diet of each), (2) 500 FTU phytase kg–1 of E. coli phytase to the positive control, and (3) the positive to the nega-

tive control. Orthogonal polynomial contrasts were used to determine if a linear or quadratic response to enzyme existed. Because of unequal spacing, appropriate coefficients were derived using the Orpol option within the Proc IML procedure of SAS. The three positive controls (one diet within each Ca series; no phytase, added dicalcium phosphate) were excluded from the data set for this analysis. Experiment 2 Diets were formulated with corn, barley, and soybean meal as the main ingredients (Table 1). Treatments consisted of a negative control diet, formulated to contain 0.44% tP (estimated 0.15% aP), and this diet supplemented with 250, 500, 1000, or 2000 FTU kg–1 of an experimental E. coli derived phytase (2600 FTU g–1; Quantum; Syngenta Animal

356

CANADIAN JOURNAL OF ANIMAL SCIENCE

Nutrition, Research Triangle Park, NC), or sufficient dicalcium phosphate to meet the aP requirement for pigs of this age (0.23%; National Research Council 1998). Growth Trial The experiment was conducted in two all-in-all-out nurseries. Each nursery consisted of 24, 1.32 m2 pens with fully slatted floors. A total of 72 barrows and 72 gilts (6.52 ± 0.75 kg, 19.6 ± 2.7 d of age; mean ± SD) were selected from two weekly farrowing outcomes. Within each nursery the pigs were randomized across treatment. The nurseries were started on test 2 wk apart to accommodate the ensuing digestibility trial. Each pen housed three pigs of the same gender. They were fed commercial Stage 1 and Stage 2 starter diets for 3 and 4 d, respectively, before commencing on experimental diets (day 0). The pre-test diets were formulated to meet the nutrient requirements for pigs of this age (National Research Council 1998), including Ca and P. The experiment lasted 28 d; pigs and feed were weighed as described in exp. 1. A blood sample (non-fasting) was collected via cranial vena cava venipuncture on day 28 from one randomly selected pig per pen approximately 4 h post-feeding. Serum was collected and frozen at –20ºC. With the exception of barrows selected for the digestibility trial (described below), all animals were euthanized and the metacarpals harvested from the left forelimb. Extraneous tissue was removed and the bones frozen at –20°C until further processing. Balance Trial Following the growth trial, 18 barrows from each room were selected to determine total tract apparent P digestibility. The medial barrow within each of the 12 pens per room containing barrows (two pens per gender per treatment), plus one other barrow that was closest to the overall treatment mean BW were selected to provide a total of three barrows per treatment per room. All pigs received the same treatment they had received during the growth trial, negating the need for an additional adaptation period. A subsample of each treatment diet used in the growth experiment was prepared with 0.4% added chromic oxide added as an indigestible marker for use in this trial. The pigs were housed individually in metabolism pens (1.2 m ± 0.56 m) which allow separate collection of faeces and urine. The pigs were weighed on day 0 (day 28 of the growth trial). Feed offered was equivalent to 3 × maintenance (110 kcal DE kg–1 BW0.75, National Research Council 1998) based on the day 0 BW and fed in two equal meals per day. Pigs were allowed 3 d to acclimatize to the room and individual housing. A 72-h collection of faeces and urine began on day 4 using a bag collection system glued around the anus (van Kleef et al. 1994) and urine collection trays under each pen. Faecal bags were changed a minimum of twice daily. Faeces were stored at – 20ºC. Excreted urine drained into a 4-L bottle containing 10 mL of 12 N HCl. The volume of urine was measured twice daily, and a 10% subsample (by volume) was stored at –20ºC. All animals were then euthanized and the metacarpal bones from the left forelimb harvested, cleaned and frozen at –20°C.

Blood Sampling and Sample Analysis A blood sample (non-fasting) was collected approximately 4 h post-feeding via cranial vena cava venipuncture from all pigs on the final day of the digestibility trial. Serum samples were submitted to a clinical chemistry lab (Prairie Diagnostic Services, Saskatoon, SK) for analysis of total Ca (colorimetrically) and inorganic P (ammonium phosphomolybdate reaction), and alkaline phosphatase (cleavage of ρ-nitrophenyl phosphate) using an automated clinical chemistry analyzer. Faeces were composited by pen and lyophilized. Feed and faeces were ground through a 1-mm screen, and analyzed for gross energy using an adiabatic bomb calorimeter (model C5000, IKA Works, Wilmington, NC) and chromic oxide (Fenton and Fenton 1979). Phosphorus in feed and faeces was partitioned into total P (tP), inorganic phosphateP, organic P and large molecule binding forms of P using a modification of a method described by Fan et al. (2001). Briefly, tP was determined colorimetrically using ammonium molybdate solution (Heinonen and Lahti 1981; Fan et al. 2001). A separate sample of faeces or feed was mixed with a 15% TCA solution and then centrifuged. The supernatant was analyzed for the TCA soluble inorganic P content and the pellet for large molecular binding forms of P using the methods described for tP. Organic P was defined as the difference between the tP and the inorganic plus large molecule binding forms of P. The large-molecule binding forms of P represent a small fraction of the insoluble P, and are thus not presented in the tables. Urinary P was analyzed by Norwest labs (Lethbridge, AB) using a modification of the ICP procedure described for feed and faeces. Thawed metacarpal bones were extracted in methanol for 24 h, diethyl ether for 48 h, dried, and then ashed at 600ºC for 16 h (Ekpe et al. 2002). The percent ash was based on the defatted dried weight of the bone. Statistical Analysis The pen (equivalent to individual in the digestibility trial) was the experimental unit. The model contained the fixed effects of treatment (growth and metabolism trials) and gender (growth trial only). Room was considered random. Week was used as a repeated measure in the growth trial. The appropriate variance-covariance matrix was chosen for each variable based on fit statistics within SAS. A linear or quadratic response to phytase was determined using contrasts. Because of unequal spacing, appropriate coefficients were derived using the Orpol option within the Proc IML procedure of SAS. The positive control treatment was excluded from the data set for this analysis. The Pdiff procedure within SAS was used to compare the negative to the positive control and 500 FTU kg–1 phytase to the positive control. Experiment 3 Treatments and Animals A 2 × 3 factorial arrangement of treatments with two levels of phytase (0 and 500 FTU kg–1diet; 2100 FTU g–1; Quantum; Syngenta Animal Nutrition, Research Triangle Park, NC) and three Ca:tP ratios was used. Diets were for-

BEAULIEU ET AL. — E. COLI DERIVED PHYTASE DECREASES TOTAL AND SOLUBLE P OUTPUT

mulated to contain 0.50, 0.80, and 1.10% Ca and 0.5% tP thus providing Ca:tP ratios of 1.0, 1.6 and 2.2 (Table 1). Limestone was added at the expense of corn to increase Ca content. Chromic oxide (0.4%) was added as an indigestible marker for the determination of digestibility. Two uniform groups of 18 barrows, born approximately 2 wk apart, were selected at weaning (7.1 ± 0.75 kg; 21.1 ± 1.0 d of age; mean ± SD) and assigned to treatment based on BW. Pigs assigned to receive a common treatment diet were housed three per pen (1.32 m2) in a nursery and fed a commercial stage 1 diet for 1 wk before commencing on experimental diets (day 0). On day 7 they were transferred into individual metabolism pens (0.66 m2). Faecal collection bags were attached on day 10 and a 5-d collection of urine and faeces began on day 11 using the collection procedures described for exp. 2. Faecal, urine and feed analysis were conducted as previously described for exp. 2. Statistical Analysis The individual animal was the experimental unit. The model contained the fixed effects of phytase enzyme, the Ca:tP ratio and their interaction. A linear or quadratic response to either phytase amount or calculated aP was determined using orthogonal polynomial contrasts. Appropriate coefficients were derived using the Orpol option within the Proc IML procedure of SAS. The Pdiff procedure within SAS was used to compare the simple effects of treatment when an interaction was detected. RESULTS General Overall health and performance were excellent. Symptoms indicative of a P deficiency were not observed. One pig refused to eat the test diet in exp. 1 and was removed from the experiment. One pig was found dead of unknown cause during the performance trial in exp. 2. No animals were removed from the trial during exp. 3. Experiment 1 This experiment used a series of 21 diets. The tP content of the 18 negative control diets, formulated to contain 0.41% tP, actually ranged from 0.33% to 0.40% tP. The average tP content of the three positive control diets was 0.53, only 0.02% less than the formulated value. Only the 0 phytase level and the positive control diet within each Ca series were analyzed for Ca. Values for the 0.50 and the 0.60% Ca diets averaged 0.52% and 0.64% respectively. Average values for the 0.70% Ca series, were higher than formulated at 0.85% and 0.76% for the negative and the positive control diets respectively (Table 1). Performance Performance results for the 28-d growth trial are shown in Table 2. Average daily gain was 1.07, 1.02, and 1.05 kg d–1 (SEM = 0.02) and average daily feed intake (ADFI) was 2.44, 2.36, and 2.49 kg d–1 (SEM = 0.05) for the 0.5, 0.6 and 0.7% Ca diets, respectively. Neither dietary Ca nor the interaction of dietary Ca and phytase level affected average daily gain (ADG), ADFI, or feed efficiency (FCE) (P > 0.10);

357

therefore, only the main effects of the phytase treatment are shown. Average daily gain improved quadratically with increasing phytase enzyme (P = 0.02, Table 2); apparently, most of the benefit was achieved at the lowest dose. Feed intake also increased in response to the first increment of enzyme (quadratic, P = 0.04) and FCE tended to improve linearly with increasing enzyme (P = 0.07). Average daily gain and ADFI of the positive control pigs (diet supplemented with dicalcium phosphate) were improved by 190 and 230 g d–1, respectively, over the negative control (P < 0.05). Supplementing the diet with 500 FTU kg–1 E. coli phytase resulted in similar ADG, ADFI and FCE compared with the diet formulated with additional dicalcium phosphate (positive control; P > 0.10). Phosphorus Digestibility Apparent P digestibility, which averaged only 21% on the negative control diet, and P output (g d–1), improved quadratically with increasing enzyme amount (P < 0.001, Fig. 1). Apparent P digestibility was not affected by enzyme source (E. coli, 39.2%, A. niger, 38.0%; P = 0.78). Neither apparent P digestibility nor faecal P output was affected by dietary Ca (P > 0.10). The interaction of phytase enzyme amount and dietary Ca was not significant for P digestibility or P output (P > 0.10). Experiment 2 Diets were formulated to contain 0.44 and 0.60% tP for the negative and positive control respectively, and 0.70% Ca. Actual tP and Ca contents were slightly higher than formulated (Table 3). The soluble inorganic fraction was approximately 60% of the total dietary P in the positive control diets, and 35 to 40% of the total dietary P in the negative control, and the negative controls with added phytase. Large molecule binding forms of P were less than 10% of tP; the remainder of the dietary P consisted of organic forms of P. Performance, Serum P and Ca, and Bone Ash Overall BW, ADG and FCE improved linearly in response to added phytase (P < 0.05; Table 4). Improved FCE was observed when the positive control was compared with either the negative control or to the 500 FTU kg–1 phytase treatment (P = 0.01). In contrast, the ADG and the ADFI were similar between the positive and the negative control treatments (P > 0.10). Serum P and Ca increased linearly in response to phytase and calculated aP in the diet (P < 0.05; Table 5). Serum P was similar between genders (P > 0.10). Examination of the data showed that serum P increased in the females in the 2000 FTU kg–1 treatment and the positive control, while in males, the overall response to phytase appeared to be linear (gender by phytase interaction, P = 0.02; Fig. 2). Serum Ca decreased linearly in response to phytase (P = 0.03). Serum Ca was 14% higher in males than females on the negative control, but almost 10% lower on the 1000 FTU kg–1 treatment (phytase by gender interaction, P < 0.05, Fig. 2). Serum alkaline phosphatase was similar between genders and phytase treatments (P > 0.10). The percent bone ash was similar between genders and tended to increase (P < 0.10) in

358

CANADIAN JOURNAL OF ANIMAL SCIENCE

Table 2. Performance response to phytase enzyme inclusion in the diet of 40 kg barrows during a 28 d trial (exp. 1)zy Phytase (U kg–1)

ADG (kg d–1)

0 (negative control) 250 x 500 x 500, A. niger x 1000 x 2000 x Positive controlw Pooled SEM P values Linear effect of enzymev Quadratic effect of enzymev Negative vs. positive control 500 FTU kg–1 (E. coli) vs. + control E.coli vs. A. niger phytase (500 FTU kg–1)

ADFI (kg d–1)

FCE

0.95 1.07 1.05 0.97 1.10 1.08 1.09 0.03

2.29 2.45 2.41 2.43 2.56 2.36 2.52 0.08

0.41 0.43 0.42 0.40 0.43 0.44 0.40 0.01

0.01 0.02 < 0.01 0.30 0.05

0.77 0.04 0.05 0.36 0.88

0.07 0.97 0.29 0.47 0.23

zMain effects of enzyme (n = 9). The effect of dietary Ca and the dietary Ca by phytase interaction were not significant. LSmeans obtained from a repeated measures analysis of weekly calculations of ADG, ADFI and FCE. yInitial BW; 40.26 ± 1.94 kg (mean ± SD). xThe phytase enzyme was added to a basal diet formulated to be limiting in aP (National Research Council 1998). Unless stated otherwise, an E.coli phytase (Quantum phytase; Syngenta Animal Nutrition, Research Triangle Park, NC) was used. wSupplemented with 0.68% dicalcium phosphate to meet available and total P requirements (National Research Council 1998). vThe positive control and the A. niger phytase treatments were removed from this analysis.

Fig. 1. The effect of phytase on tP digestibility (%) and output (g d–1; exp. 1, n = 9). Effect of phytase on tP digestibility (analysis excluded the positive control); linear, P < 0.0001, quadratic, P < 0.001; SEM = 2.33. Effect of phytase on P output (analysis excluded the positive control) linear, P < 0.001, quadratic, P = 0.07; SEM = 0.26. Effects of dietary Ca, and all interactions, P > 0.10 for both digestibility and output (exp. 1; n = 9).

response to phytase (linearly). Compared with the positive control diet, the serum Ca was greater when pigs consumed the negative control diet (P < 0.01); however, the opposite response was observed with serum P and percent bone ash (P < 0.02). Serum P was higher when pigs consumed the diet supplemented with added dicalcium phosphate (positive control) compared with that observed with the addition of 500 FTU phytase kg–1 (P < 0.01), while the opposite response was observed with serum Ca and alkaline phosphatase (P < 0.01; Table 5). Energy digestibility was not affected by treatment (P > 0.10; Table 6). Although diets were formulated to be similar in tP and feed intake did not differ small differences in the actual tP content of the diet and changes in feed intake combined to result in increased tP intake as the phytase content of the diet increased (P < 0.05; Table 6). Soluble inorganic

P intake was greatest at the extremes of phytase supplementation (quadratic, P < 0.05). The intake of P was greatest when pigs consumed the positive control diet, relative to either the negative control (negative vs. positive control, P < 0.05) or the 500 FTU kg–1 phytase treatment (500 FTU phytase vs. positive control, P < 0.05). Apparent digestibility of tP was greatest with the addition of 1000 FTU kg–1 of phytase enzyme (response to phytase, quadratic, P < 0.05; Table 6). Apparent digestibility of tP was higher in the positive control, relative to the negative control (negative vs. positive control, P < 0.05) and was similar between the positive control and the 500 FTU kg–1 phytase treatment (P > 0.10). Apparent digestibility of the soluble inorganic P fraction, which was negative with 0, 250 or 500 U kg–1 of phytase addition increased quadratically in response to phytase enzyme (P < 0.05). Apparent digestibility of this fraction was much higher in the positive control treatment relative to either the negative control or the 500 FTU kg–1 phytase treatment that was formulated to be equal to the positive control in calculated aP (P < 0.05, negative vs. positive control; 500 FTU kg–1 vs. positive control). Apparent digestibility of the organic P fraction was not affected by phytase (P > 0.10). The faecal excretion of the various fractions of P, which were calculated from the corresponding digestibility coefficients, followed similar trends as digestibility (Table 6). Excretion of total and the soluble inorganic fraction of P responded quadratically, in response to phytase (P < 0.05). The excretion of the organic fraction of P, which never exceeded 0.50 g pig–1 d–1, was unaffected by dietary phytase (P > 0.10). The excretion of P in the urine, which was dramatically increased on the positive control diet, increased linearly in response to phytase (P < 0.05). Faecal excretion of the total and the soluble inorganic fraction of P was greatest when the diet was supplemented with an inorganic source of P (negative vs. positive control and 500 FTU kg–1 phytase vs. the positive control, P < 0.05). Expressed as a proportion of the percent of tP intake, the soluble inor-

BEAULIEU ET AL. — E. COLI DERIVED PHYTASE DECREASES TOTAL AND SOLUBLE P OUTPUT

359

Table 3. Analyzed phosphorus and calcium content (% as fed) of experimental dietsz (exp. 2) Phytase (U kg–1) aPy

0 0.15

250 0.22

500 0.29

1000 0.43

2000 0.44

0 0.31

Total P (%) Soluble inorganic P Large-molecule binding forms of P Organic Px Calcium (%)

0.52 0.20 0.05 0.28 0.95

0.52 0.18 0.05 0.29 0.82

0.52 0.18 0.04 0.30 0.82

0.55 0.18 0.05 0.33 0.87

0.51 0.19 0.05 0.27 0.84

0.70 0.42 0.05 0.23 0.77

zBased on the method of Fan et al. (2001). yCalculated available P, assuming 100 FTU of E. coli derived enzyme improved xCalculated; organic P = total P – (soluble inorganic P + large molecule-binding

P availability by 0.028 percentage points. forms of P).

Table 4. Performance response of weanling pigs to increasing phytase concentration in a diet based on barley, corn and soybean meal (exp 2)zy Treatments Phytase (U kg–1)

aPx (%)

0 (NC)w 250 500 1000 2000 0 (PC)w SEM P values

0.15 0.22 0.29 0.43 0.44 0.31 Phytase, linearv NC vs. PC 500 FTU phytase vs. PC

ADG (kg d–1)

ADFI (kg d–1)

FCE, gain:feed

0.44 0.45 0.46 0.47 0.47 0.47 0.02 0.03 0.19 0.51

0.66 0.67 0.68 0.67 0.69 0.66 0.10 0.87 0.95 0.94

0.66 0.67 0.66 0.68 0.69 0.72 0.01 0.01 0.01 0.01

zLSMeans,

four pens per gender per treatment, three pigs per pen. Main effects of treatment only. The effect of gender and the gender by treatment interaction were not significant. LSMeans shown were obtained from a repeated measures analysis of weekly means. yInitial BW = 7.36 ± 0.80 kg (mean ± SD). xCalculated available P, assuming 100 FTU of phytase enzyme (Quantum; Syngenta Animal Nutrition, Research Triangle Park, NC) improves total P availability by 0.028 percentage points. wNC (negative control), 0.44% tP. PC (positive control); dicalcium phosphorous added to the diet to increase tP to 0.60% and fulfil the aP requirement for pigs of this age (NRC 1998) vLinear response to increasing phytase. The positive control was not included in this analysis. Quadratic, P > 0.10. Table 5. Effects of phytase on serum Ca, P, alkaline phosphatase, and bone ash of weanling pigs (exp 2)z Serumy

Treatments Phytase (U kg–1)

aPx (%)

0 (NC)w 250 500 1000 2000 0 (PC)w SEM P values

0.15 0.22 0.29 0.43 0.44 0.31 Phytase, linearv Phytase by gender NC vs. PC 500 U phytase vs. PC

P (mmol L–1)

Ca (mmol L–1)

Alk. P (U L–1)

Bone ash (%)

2.46 2.64 2.67 2.71 3.03 3.13 0.09 < 0.001 0.02 < 0.001 < 0.001

3.16 3.20 3.27 3.08 3.00 2.91 0.07 0.03 0.03 0.01 < 0.001

323 338 392 330 324 301 22.8 0.57 0.37 0.50 0.008

40.1 41.5 42.2 40.8 43.3 43.2 2.0 0.06 0.63 0.02 0.45

zn = 6. Effect of gender, P > 0.10. ySerum collected from 3 pens per treatment per block. Block 1, two pigs per pen; block 2, one pig per pen. xCalculated available P, assuming 100 FTU of phytase enzyme (Quantum; Syngenta Animal Nutrition, Research

Triangle Park, NC) improves total P availability by 0.028 percentage points. wNC (negative control), limiting in aP. PC (positive control); sufficient dicalcium phosphorous added to the diet to fulfil the aP requirement for pigs of this age (NRC 1998) vLinear response to increasing phytase. The positive control was not included in this analysis. Quadratic, P > 0.10.

ganic fraction of P output was greater in the positive control relative to the negative control (negative vs. positive control, P < 0.05). Experiment 3 Actual tP content of the diets ranged from 0.50 to 0.53%,

only slightly higher than the 0.50% formulated values. The Ca content of the diets formulated to contain 0. 50% Ca actually ranged from 0.54 to 0.57% Ca. Similar values for the diets formulated to contain 0.80 and 1.10% Ca were 0.85 to 0.86% and 1.15 to 1.20%. The actual Ca:tP ratios were within 0.08% of formulated.

360

CANADIAN JOURNAL OF ANIMAL SCIENCE

was similar regardless of enzyme (phytase by Ca:tP ratio interaction, P = 0.004). Response to Dietary Ca:tP Ratio The dietary Ca:tP ratio had no effect on the apparent digestibility of energy, Ca, or the soluble inorganic fraction of P (P > 0.10; Table 7). Digestibility of the organic fraction of P decreased linearly as the Ca:tP ratio increased (P < 0.05). An increasing Ca:tP ratio resulted in a linear increase in the output of Ca, tP, and the organic fraction of P (P < 0.05). DISCUSSION

Fig. 2. The effect of phytase on serum P and Ca in male and female weanling pigs (Exp. 2; n=6). Phosphorus and Ca responded in a linear fashion to phytase, P < 0.05. Diet by gender interaction for P and Ca, P < 0.05.

Main Effects of Phytase Enzyme and the Interaction with the Ca:tP Ratio An interaction of the phytase enzyme by the diet Ca:tP ratio was always observed when the effect of phytase was significant. Supplementing the diet with 500 FTU kg–1 phytase had no effect on apparent energy or Ca digestibility (P > 0.10; Table 7). Apparent P digestibility was greatest at the two lower Ca:tP ratios and when the diets were supplemented with 500 FTU kg–1 phytase (phytase by Ca:tP ratio interaction, P = 0.02). Apparent digestibility of the soluble inorganic forms of P increased from –9.5 to 10.3% (P < 0.001) with the addition of 500 FTU kg–1 phytase. Conversely, only a 5% improvement in apparent digestibility was observed in the soluble organic fraction of P with enzyme addition (84.2 to 88.6%, P > 0.10). The highest apparent digestibility of the soluble inorganic fraction of P was observed on the intermediate Ca:tP ratio and with added phytase in the diet (phytase by Ca:tP ratio interaction, P = 0.001). Intake of tP was similar between diets (P > 0.10) and while the intake of the soluble inorganic and organic fractions of P differed between diets, a common trend is difficult to detect (phytase by Ca:tP ratio, P < 0.01). Excretion of total and the soluble inorganic fraction of P in the faeces was increased when the pigs received phytase in their diets, except at the highest Ca:tP ratio when the excretion was not affected by the enzyme (phytase by Ca:tP ratio interaction, P = 0.009). Conversely, the excretion of the organic forms of P, which accounted for approximately 15% of the excreted P, was similar regardless of enzyme addition (P > 0.05). Urinary excretion of P, which averaged less than 1% of the total P excreted, was higher in the pigs at the lowest dietary Ca:tP ratio and when phytase was included in the diet (phytase by Ca:tP ratio interaction, P = 0.001). The output of soluble inorganic P output, expressed as a proportion of total P intake, was greatest when phytase was not included in the diet, except at the highest Ca:tP ratio, when this proportion

Performance The primary motivation for the use of phytase enzymes in monogastric nutrition has been to improve the utilization of phytate-P in feedstuffs. Potential benefits therefore include a reduction in the tP and inorganic P content of diets resulting in decreased output of P in the manure. However, for commercial acceptance, it is also important that performance be maintained relative to a diet containing sufficient aP. The diets in these experiments were formulated with negative and positive controls designed to provide inadequate and adequate amounts of P, respectively, based on National Research Council (1998) estimations of requirements. The negative control was formulated so that supplementation of this diet with 500 FTU kg–1 phytase enzyme would meet the aP requirements for pigs of the appropriate age category based on an estimate of the improvement in available P of 0.028% per 100 FTU of enzyme. This estimate is similar to the P release value of 0.108% per 400 FTU kg–1 observed by Augspurger et al. (2003) using an experimental E. coli phytase enzyme in the diet of young pigs. However, as shown in exp. 1, linearity between phytase content of the diet and P digestibility can not be assumed. Performance was not measured in the third experiment; however, in exp. 1 a significant improvement in growth and, in exp. 2, an improvement in FCE with no change in feed intake, when the positive is compared with the negative control suggest that the negative control was indeed deficient in aP. Performance responses to phytase, however, were variable between experiments. In the first experiment, a modest performance response, improved ADG and FCE was observed. Similarly, in the second experiment, which utilized younger pigs receiving approximately 80% of their tP requirements (National Research Council 1998), the performance response to phytase was also modest. The improvement in exps. 1 and 2, with the addition of 500 FTU phytase to a diet containing either 0.37% tP (40 kg BW pigs) or 0.44% tP (7.3 kg BW pigs) averaged approximately 20 g d–1. This contrasts with the data of Veum et al. (2006) who observed increases in growth of 60 g d–1 when 500 FTU phytase was added to a corn–soybean meal diet formulated to contain [Author?] than 0.45% tP and fed to 7.6 kg pigs for 28 d. The plasma P and the percent bone ash data substantiate our conclusion that the negative control diets were indeed limiting in tP. Plasma P and the percent bone ash

BEAULIEU ET AL. — E. COLI DERIVED PHYTASE DECREASES TOTAL AND SOLUBLE P OUTPUT

361

Table 6. The effect of phytase on apparent energy and P digestibility and output by male weanling pigs (exp 2)z Phytase (U kg–1) aPy Energy digestibility (%) P intake (g d–1) Totalw Soluble inorganicw,v Organicv P apparent digestibility (%) Totalw,v Soluble inorganicw,v Organicw,v P output Faecal (g d–1) Totalw,v Soluble inorganicw,v Organicv Urinary P (mg d–1)w P retained (g d–1)w Soluble inorganic P output (% of total intake)w

0 0.15 79.5 5.03 1.88 2.68 40.1 –27.8 86.8 3.01 2.40 0.35 8.7 2.01 47.7

250 0.22

500 0.29

1000 0.43

2000 0.44

80.9

80.3

81.5

79.9

5.02 1.76 2.78 46.9 –19.4 87.8 2.68 2.11 0.34 10.8 2.33 41.9

4.91 1.66 2.83 51.8 –17.4 93.3 2.36 1.95 0.19 6.6 2.28 39.7

5.17 1.68 3.04

5.27 1.95 2.81

0 0.31 79.6 7.19 4.31 2.37

SEM

Regx

0.84

NS

0.30 0.13 0.15

L L,Q Q

59.7 5.3 91.0

55.7 7.6 89.2

47.8 30.8 80.4

2.1 4.8 3.2

L,Q L,Q NS

2.09 1.60 0.28 18.1 3.06 30.8

2.33 1.80 0.30 21.8 2.92 34.2

3.75 2.98 0.46 208.9 3.23 41.5

0.22 0.18 0.09 19.1 0.13 1.7

L,Q L,Q NS L L, Q L,Q

zn = 6. Initial BW = 7.62 ± 0.81 kg (mean ± SD) yCalculated available P, assuming 100 FTU kg–1

of phytase enzyme (Quantum phytase; Syngenta Animal Nutrition, Research Triangle Park, NC) improves P availability by 0.028 percentage points. xL, Q, linear or quadratic response to increasing phytase, P < 0.05; the positive control was not included in this analysis. wContrast, negative vs. positive control, P < 0.05. vContrast 500 FTU phytase kg–1 vs. positive control, P < 0.05.

were lower in pigs receiving the negative relative to the positive control diet. Ekpe et al. (2002) observed quadratic responses of alkaline phosphatase, plasma P and metacarpal bone ash in response to increasing dietary digestible P. Their calculated P requirement was slightly lower with plasma P (6.01 g d–1) than ADG (6.17 g d–1). Lei et al. (1993a) and Stahl et al. (2000) concluded that plasma P concentration was the most sensitive measure of the effect of the phytase enzyme on phytate-P utilization and was correlated with bone strength. Nutrient Digestibility Energy Energy digestibility, not measured in exp. 1, was unaffected by phytase in exps. 2 or 3. Although, increases in energy (Kies et al. 2005), starch and dry matter digestibility (Johnston et al. 2004) in response to 500 to 1000 FTU phytase kg–1 diet have been observed, the improvement is typically only 1 to 2 percentage units. Phosphorus Total Phosphorus Digestibility. Apparent P digestibility in exps. 2 and 3 by pigs consuming diets supplemented with 500 FTU kg–1 of E. coli derived phytase averaged 51.8 and 53.3%, respectively. Apparent P digestibility in exp. 3, with 500 FTU kg–1 phytase and a low dietary Ca:tP ratio, averaged 57.5%. The incremental improvement in exps. 2 and 3 with the addition of 500 FTU phytase to the basal diet was similar, (11.7% in exp. 2 and 12.1% in exp. 3). This resulted in a reduction in tP output in the faeces plus urine of about 22% in both of these experiments. Conversely, in exp. 1, tP digestibility on the basal diet was only about 20%, and the incremental improvement following the supplementa-

tion of 500 FTU phytase kg–1 was almost 86%. The diets were based on corn and soybean meal in exp. 1. and on corn, soybean meal and barley in exps. 2 and 3. There is some evidence indicating differences between feedstuffs regarding inorganic phosphate release in response to supplemental microbial phytase. For example, when several studies were summarized, Johansen and Poulsen (2003) calculated an average increase in digestible P content (g kg–1 diet) of 0.26 and 0.44 in wheat/barley or corn-based diets, respectively, with the addition of 250 FTU phytase kg–1 diet. The difference between feedstuffs became less as increasing amounts of phytase were supplemented to the diets (Johansen and Pulsen 2003). In vitro studies using an E. coli derived phytase showed that inorganic phosphate release was higher for soybean meal relative to corn (Adeola et al. 2004). Our diets in exp. 1 contained a higher proportion of corn relative to soybean meal than was used in exp. 2. Urinary P output, measured only in exps. 2 and 3, was variable and represented a small fraction of total P output. This has been observed by others when diets were fed to weanling pigs without supplemental P (Lei et al. 1993a). Phosphorus output in the urine was consistently less than 1% of total P output. Exceptions are the positive control in exp. 2, where output was 5.6% of the total output, and the low Ca:tP ratio plus phytase treatment in exp. 3, where the output was 3.7% of the total output. Similarly, when pigs (21 to 60 kg) were fed diets with Ca:P ratios ranging from 1.5 to 2.2, and plus or minus phytase, P output in the urine ranged from 29 to 38 mg L–1, except on the low Ca:P ratio diet with added phytase, where P output in the urine was 205 mg L–1 (Seynaeve et al. 2000). Ekpe et al. (2002) showed that urinary P, as a percent of total output, increased by approximately 3-fold between diets which were deficient to adequate in digestible P.

362

CANADIAN JOURNAL OF ANIMAL SCIENCE

Table 7. Simple effects of apparent energy and phosphorus digestibility and phosphorus balance in weanling pigs fed diets with or without 500 U kg–1 phytase enzyme and increasing Ca:tP ratioz (exp. 3) Ca:tP ratioy Phytase (U

1.12 kg–1)

Apparent digestibility (%) Energy Calcium Phosphorus Total Soluble inorganic Organic Ca Intake (g d–1) P Intake (g d–1) Total Soluble inorganic Organic Ca output (g d–1) P output Faecal Total (g d–1 Soluble inorganic (g d–1) Organic (g d–1) Urinary (mg d–1) P retained (g d–1) Soluble inorganic P output (% of total P intake)

1.66

2.31

P values

0

500

0

500

0

500

SEM

Phytase

Ca:tPx

83.1 68.7

82.4 71.5

82.3 62.6

82.3 71.5

83.7 61.6

81.9 65.9

0.8 4.2

0.2 0.1

NSw NS

0.57 0.76

45.1bc –1.7bc 87.1 2.88d

57.2a 8.1b 93.7 2.69d

55.1a 24.4a 88.7 4.17c

43.9bcd –5.2c 79.5 5.54a

47.7b –1.7bc 83.3 6.16b

3.4 7.8 3.2 0.21

0.001 0.001 0.10 0.44

L NS L L

0.02 0.001 0.83 0.02

2.50 0.96bc 1.13bc 1.38

2.62 1.17a 0.81e 1.01

2.38 0.86d 1.17b 1.91

2.69 0.96bc 1.30a 1.81

0.13 0.05 0.05 0.18

0.11 0.11 0.10 0.24

NS Q L,Q L

0.11 0.001 0.001 0.67

1.54a 1.17a 0.16 3.2a 2.50

1.18b 0.89cd 0.09 2.8a 2.62

1.34ab 0.91cd 0.24 2.7a 2.38

1.42a 0.97bc 0.22 1.8a 2.69

0.14 0.11 0.04 4.6 0.13

0.002 0.001 0.13 0.002 0.11

L NS L L,Q NS

0.009 0.004 0.77 0.001 0.11

3.0

0.001

NS

0.004

2.52 1.05ab 1.02d 0.72 1.39ab 1.06ab 0.13 9.2a 2.52 42.1a

2.45 0.87cd 1.06cd 0.65 1.05c 0.81d 0.07 39.8b 2.45 32.9d

38.2d –21.5d 86.0 4.31c

46.9c

33.8d

37.9bc

36.2cd

Interaction

a–d Means within a row with different letters are different (P < 0.05). zn = 6 individually housed barrows per treatment in two blocks (7.1 ± 0.75 kg; mean ± SD). yBased on analyzed mean values for tP, and averaged across phytase treatments. xL, linear effect of Ca:tP ratio (P < 0.05), Q, quadratic effect of Ca:tP ratio (P < 0.05). wNS, not significant, (P > 0.10).

Effect of Phytase on the Form of P Excreted. Phytase releases orthophosphate groups from the phytate molecule, and it has been suggested that phytase may increase the output of soluble P. For example, when Daumer et al. (2004) fractionated the P in swine manure into six different fractions ranging in solubility, they found that while the tP concentration was almost 12% lower in the slurry from farms where phytase was used, none of this P was in the most insoluble form. In a recent review, Maguire et al. (2005) compared three studies examining the use of phytase in swine diets. Total P in the manure decreased by 7 to 27% with the addition of phytase; however, the proportion of tP that was water soluble actually increased in two of the three studies. Different methodologies have been used to estimate water-soluble P in manure and the propensity to leach into surrounding water. We used the method described by Fan et al. (2000, 2001), which uses a 15% trichloroacetic acid solution to partition tP into fractions described as the soluble inorganic and soluble organic P. According to Fan et al. (2000), the soluble inorganic P is the most readily available form of P. Apparent digestibility of the soluble inorganic fraction of P was negative on the negative control diets in exps. 2 and 3 and when 250 and 500 FTU phytase kg–1 diet was added in exp. 2, indicating a greater faecal output than dietary intake of inorganic P. This was unexpected, but could be because the P released exceeded intestinal absorptive capacity or, as shown by Fan et al. (2001) due to phytase activity in the large intestine. Many factors control P homeostasis within the body (reviewed by Jongbloed 1987). In exp. 2, as

the phytase supplementation increased, the apparent digestibility of tP and the soluble inorganic fraction of P improved as did P retention. This indicates that as P availability improved, the pig absorbed and retained proportionally more P. In exps. 2 and 3, approximately 80% of the faecal tP was soluble inorganic P. This is higher than the 47% observed by Fan et al. (2000) in representative samples from weaning pigs fed commercial diets and is difficult to reconcile. The addition of 500 FTU phytase kg–1 to a low P basal diet in exp. 2 decreased the output of soluble inorganic P from approximately 2.4 g pig–1 d–1 to 2.0 g pig–1 d–1 in 20-kg pigs. Output on the positive control diet (added dicalcium phosphate) was almost 3 g pig –1 d–1. The amount of soluble inorganic P, as a proportion of tP, decreased from 48 to 40% with the addition of 500 FTU kg–1 and further to 31% with 1000 FTU kg–1. Similar results were seen in the younger pigs (exp. 3). Contradicting their work in broilers (Angel et al. 2005), the proportion of tP that was water-soluble was not affected by the addition of 500 FTU kg–1 phytase to pig diets formulated to meet National Research Council requirements for tP, or a diet with reduced phytate P (Angel et al. 2005). The authors suggested that the reduced P diet contained more P than the animals required. They concluded that when phytase is added to a diet that contains no excess P, then both total and water-soluble output will be decreased. This is confirmed by our results. Effect of Ca:tP Ratio. In our third experiment, 500 FTU kg–1 phytase enzyme increased apparent tP digestibility by 22% when the dietary Ca:tP ratio was 1.1, but by only 8% when the Ca:tP ratio had increased to 2.3. Conversely, although

BEAULIEU ET AL. — E. COLI DERIVED PHYTASE DECREASES TOTAL AND SOLUBLE P OUTPUT

the effect of phytase on P digestibility was greater in exp. 1, regardless of pig age/weight, enzyme source, or level of enzyme addition, there was no effect of the dietary Ca:tP ratio on phytase efficacy. Examination of the interaction between Ca:tP ratios and phytase in exp. 1 must proceed cautiously, as replication of the simple effects was only with three pigs. Adverse effects of a wide dietary Ca:tP ratio on phytase efficacy are well documented (Qian et al. 1996; Brady et al. 2002) and the lack of an effect in exp. 1 was unexpected. In both experiments, diets were formulated to be adequate, or slightly limiting in tP, and the Ca:tP ratio was increased by increasing the Ca concentration in the diet with added limestone at a constant P concentration. The animals used in exp. 1 were older. Lei et al. (1994) however, found a depressive effect of dietary Ca level on phytase activity in weanling pigs, and Liu et al. (1998) showed that P utilization in growing-finishing pigs in the presence of phytase was improved at a Ca:P ratio of 1 rather than 1.5, indicating that age does not affect this response. The inhibitory effect of Ca on phytase efficacy was shown to be less at pH 2.5 than at pH 6.5; probably because of decreased solubility of the complex at the higher pH (Tamin et al. 2004). The single pH optima of 2.5 for E. coli derived phytases is lower than the optima for other phytases (Adeola et al. 2004); therefore, it could be expected that the inhibitory effect of Ca would be less. Unfortunately, gastrointestinal tract pH was not determined in our studies therefore we could not test this hypothesis. In conclusion, the supplementation of swine diets with 500 to 1000 FTU kg–1 E. coli derived phytase enzyme decreases tP output in the manure by 30 to 50% regardless of age of the pig. The excretion of soluble P in the manure is similarly decreased, provided the diets do not contain excessive total P. Excessive dietary Ca had a detrimental effect on phytase efficacy. We observed decreases in phytase efficacy when the dietary Ca:tP ratio was between 1.66 and 2.31. ACKNOWLEDEGMENTS Financial support from Syngenta Animal Nutrition, Research Park Triangle, NC, is gratefully acknowledged. Strategic funding is provided to PSCI by Sask Pork, Alberta Pork, Manitoba Pork Council, and Saskatchewan Agriculture and Food Development Fund. Adeola, O., Sands, J. S., Simmins, P. H. and Schulze, H. 2004. Efficacy of an Escherichia coli-derived phytase preparation. J. Anim. Sci. 82: 2657–2666. Ajakaiye, A., Fan, M. Z., Archbold, T., Hacker, R. R., Forsberg, C. W. and Phillips, J. P. 2003. Determination of true digestive utilization of phosphorus and the endogenous phosphorus outputs associated with soybean meal for growing pigs. J. Anim. Sci. 81: 2766–2775. American Oil Chemists’ Society. 1993. Official methods and recommended practices. 4th ed. AOCS, Champaign, IL. Angel, C. R., Powers, W. J., Applegate, T. J., Tamin, N. M. and Christman, M. C. 2005. Influence of phytase on water-soluble phosphorus in poultry and swine manure. J. Environ. Qual. 34: 563–571. Association of Official Analytical Chemists. 1995. Official

363

methods of analysis. 17th ed. AOAC, Arlington, VA. Augspurger, N. R., Webel, D. M., Lei, X. G. and Baker, D. H. 2003. Efficacy of an E. coli phytase expressed in yeast for releasing phytate-bound phosphorus in young chicks and pigs. J. Anim. Sci. 81: 474–483. Brady, S. M., Callan, J. J., Cowan, D., McGrane, M. and O’Doherty, J. V. 2002. Effect of phytase inclusion and calcium/phosphorus ratio on the performance and nutrient retention of grower-finisher pigs fed barley/wheat/soya bean meal-based diets. J. Sci. Food Agric. 82: 1780–1790. Canadian Council on Animal Care. 1993. Guide to the care and use of experimental animals. E. D. Olfert, B. M. Cross, and A. A. McWilliam, eds. CCAC, Ottawa, ON. Daumer, M.-L., Beline, F. and Sperandio, M. 2004. The effect of phytase in pig diets and solid/liquid separation of pig slurry on phosphorus, calcium, and magnesium fractionation. Trans. ASAE 47: 1247–1253. DeLaune, P. B., Moore, P. A., Jr., Carman, D. K., Sharpley, A. N., Haggard, B. E. and Daniel, T. C. 2004. Development of a phosphorus index for pastures fertilized with poultry litter-factors affecting phosphorus runoff. J. Environ. Qual. 33: 2183–2191. Eeckhout, W. and De Paepe, M. 1994. Total phosphorus, phytate-phosphorus and phytase activity in feedstuffs. Anim. Feed. Sci. Technol. 47: 19–29. Engelen, A. J., van der Heeft, F. C., Randsdorp, P. H., Somers, W. A., Schaefer, J. and van der Vat, B. J. 2001. Determination of phytase activity in feed by a colorimetric enzymatic method: collaborative interlaboratory study. J. Am. Oil Chem. Int. 84: 629–633. Ekpe, E. D., Zijlstra, R. T. and Patience, J. F. 2002. Digestible phosphorus requirement of grower pigs. Can. J. Anim. Sci. 82: 541–549. Fan, M. Z., Archbold, T., Mead, K. and Gao, Y. 2000. Partitioning various forms of phosphorus in the distal ileal digesta and feces. Can. J. Anim. Sci. 80: 780 (Abstr.). Fan, M. Z., Archbold, T., Sauer, W. C., Lackeyram, D., Rideout, T., Gao, Y., de Lange, C. F. M. and Hacker, R. R. 2001. Novel methodology allows simultaneous measurement of true phosphorus digestibility and the gastrointestinal endogenous phosphorus outputs in studies with pigs. J. Nutr. 131: 2388–2396. Fenton, T. W. and Fenton, M. 1979. An improved procedure for the determination of chromic oxide in feed and feces. Can. J. Anim. Sci. 59: 631–634. Greiner, R., Farouk, A., Alminger, M. L. and Carlsson, N.-G. 2002. The pathway of dephosphorylation of myo-inositol hexakisphosphate by phytate-degrading enzymes of different Bacillus spp. Can. J. Microbiol. 48: 986–994. Heinonen, J. K. and Lahti, R. J. 1981. A new and convenient colorimetric determination of inorganic orthophosphate and its application to the assay of inorganic pyrophosphate. Anal. Biochem. 133: 313–317. Jendza, J. A., Dilger, R. N., Adedokum, S. A., Sands, J. S. and Adeola, O. 2005. Escherichia coli phytase improves growth performance of starter, grower, and finisher pigs fed phosphorus-deficient diets. J. Anim. Sci. 83: 1882–1889. Johnston, S. L., Williams, S. B., Southern, L. L., Bidner, T. D., Bunting, L. D., Matthews, J. O. and Olcott, B. M. 2004. Effect of phytase addition and dietary calcium and phosphorus levels on plasma metabolites and ileal and total-tract nutrient digestibility in pigs. J. Anim. Sci. 82: 705–714. Jongbloed, A. W. 1987. Phosphorus in the feeding of pigs. Ph.D. thesis, Institute voor Veevoedingsonderzoek, Lelystad, the Netherlands. Johansen, K. and Poulsen, H. D. 2003. Substitution of inorganic

364

CANADIAN JOURNAL OF ANIMAL SCIENCE

phosphorus in pig diets by microbial phytase supplementation – a review. Pig News Info. 24: 77N–82N. Kies, A. K., Gerrits, W. J. J., Schrama, J. W., Heetkamp, M. J. W., van der Linden, K. L., Zandstra, T. and Verstegen, M. W. A. 2005. Mineral absorption and excretion as affected by microbial phytase and their effect on energy metabolism in young piglets. J. Nutr. 135: 1131–1138. Lei, X. G., Ku, P. K., Miller, E. R. and Yokoyama, M. T. 1993a. Supplementing corn-soybean meal diets with microbial phytase linearly improves phytate phosphorus utilization by weanling pigs. J. Anim. Sci. 71: 3359–3367. Lei, X. G., Ku, P. K., Miller, E. R., Yokoyama, M. T. and Ullrey, D. E. 1993b. Supplementing corn-soybean meal diets with microbial phytase maximizes phytate phosphorus utilization by weanling pigs. J. Anim. Sci. 71: 3368–3375. Lei, X. G., Ku, P. K., Miller, E. R., Yokoyama, M. T. and Ullrey, D. E. 1994. Calcium level affects the efficacy of supplemental microbial phytase in corn-soybean meal diets of weanling pigs. J. Anim. Sci. 72: 139–143. Leytem, A. B., Mikkelsen, R. L. and Gilliam, J. W. 2002. Sorption of organic phosphorus compounds in Atlantic coastal plain soils. 167: 652–658. Liu, J., Bollinger, D. W., Ledoux, D. R. and Veum, T. L. 1998. Lowering the dietary calcium to total phosphorus ratio increases phosphorus utilization in low-phosphorus corn-soybean meal diets supplemented with microbial phytase for growing-finishing pigs. J. Anim. Sci. 76: 808–813. Maguire, R. O., Dou, Z., Sims, J. T., Brake, J. and Joern, B. C. 2005. Dietary strategies for reduced phosphorus excretion and improved water quality. J. Environ. Qual. 34: 2093–2103. National Research Council. 1998. Nutrient requirements of swine. 10th rev. ed. National Academy Press, Washington, DC.

Qian, H., Kornegay, E. T. and Conner, D.E. Jr. 1996. Adverse effects of wide calcium:phosphorus ratios on supplemental phytase efficacy for weanling pigs fed two dietary phosphorus levels. J. Anim. Sci. 74: 1288–1297. Rodriguez, E., Han, Y. and Lei, X. G. 1999. Cloning, sequencing, and expression of an Escherichia coli acid phosphatase/phytase gene (appA2) isolated from pig colon. Biochem. Biophys. Res. Commun. 257: 117–123. Seynaeve, M., Janssens, G., Hesta, M., van Nevel, C. and De Wilde, R. O. 2000. Effects of dietary Ca/P ratio, P level and microbial phytase supplementation on nutrient digestibilities in growing pigs: precaecal, post-ileal and total tract disappearances of OM, P and Ca. J. Anim Physiol. Anim. Nutr. 83: 36–48. Simon, O. and Igbasan, F. 2002. In vitro properties of phytase from various microbial origins. Int. J. Food Sci Technol. 37: 813–822. Stahl, C. H., Roneker, K. R., Thornton, J. R. and Lei, X. G. 2000. A new phytase expressed in yeast effectively improves the bioavailability of phytate phosphorus to weanling pigs. J. Anim. Sci. 78: 668–674. Tamin, N.M., Angel, R. and Christman, M. 2004. Influence of dietary calcium and phytase on phytate phosphorus hydrolysis in broiler chickens. Poultry Sci. 83: 1358–1367. van Kleef, D. J., Deuring, K. and van Leeuwen, P. 1994. A new method of faeces collection in the pig. Lab. Anim. 38: 78–79. Veum, T.L., Bollinger, D.W., Buff, C.E. and M.R. Bedford. 2006. A genetically engineered Escherichia coli phytase improves nutrient utilizaiton, growth performance, and bone strength of young swine fed diets deficient in available phosphorus. J. Anim. Sci. 84: 1147–1158.