Determining the phosphorus release of GraINzyme phytase in diets for nursery pigs

Abstract

The objective of this study was to determine the available P (aP) release curve for a new phytase source, GraINzyme Phytase (Agrivida Inc., Woburn, MA), which is expressed in corn containing an engineered Escherichia coli phytase called Phy02. Plant-expressed phytases are created by inserting phytase-encoding genes into plants resulting in their ability to produce seeds with increased concentrations of phytase. A total of 360 pigs (Line 200 × 400, DNA, Columbus, NE, initially 9.9 ± 0.19 kg) were used in a 21-d growth study. Pigs were weaned at approximately 21 d of age, randomly allotted to pens based on initial body weight (BW) and fed common starter diets. From days 18 to 21 postweaning, all pigs were fed a diet containing 0.11% aP. On day 21 postweaning, considered day 0 of the study, pens were blocked by BW and randomly allotted to one of eight dietary treatments with five pigs per pen and nine pens per treatment. Dietary treatments were formulated to include increasing aP derived from either an inorganic P source (0.11%, 0.19%, or 0.27% from monocalcium P) or increasing phytase (150, 250, 500, 1,000, or 1,500 FTU/kg). Diets were corn-soybean meal-based and contained 1.24% standardized ileal digestible Lys. On day 21 of the trial, one pig per pen (weighing closest to the mean pen BW) was euthanized and the right fibula was collected to determine bone ash using the nondefatted processing method. Overall (days 0 to 21), pigs fed increasing aP from inorganic P or phytase had increased (linear, P < 0.002) average daily gain (ADG), average daily feed intake (ADFI), and gain-to-feed (G:F; quadratic, P < 0.05). Bone ash weight (g) and percentage bone ash increased (linear, P < 0.001), with increasing inorganic P or added phytase. Based on the composition of the diets used in this study, the release equations developed for GraINzyme for ADG, G:F, bone ash weight, and percentage bone ash are as follows: aP = (0.255 × FTU)/(1299.969 + FTU), aP = (0.233 × FTU)/(1236.428 + FTU), aP = (45999.949 × FTU)/(462529200 + FTU), and aP = (0.272 × FTU)/(2576.581 + FTU), respectively.

INTRODUCTION

Most swine diets are formulated using ingredients that contain phytate-bound phosphorus. Phytic acid or phytate is a sixfold dihydrogen phosphate ester of inositol that is the major storage form (60% to 82%) of P in grains (Ravindran et al., 1994). Monogastric species do not naturally synthesize the enzyme phytase to cleave the phosphates from the phytic acid for absorption (Selle and Ravindran, 2007). Therefore, exogenous phytase can be added to swine diets to make P more available for growth and bone development (Cromwell et al., 1993; Bento et al., 2012; Rutherfurd et al., 2014). The added phytase allows for reduced dietary additions of inorganic P, which results in reduced P excretion and lower feed cost (Selle and Ravindran, 2007; Li et al., 2013).

Phytase was first commercialized in 1991 and is known to be one of the most significant discoveries in animal nutrition (Cromwell, 2009; Li et al., 2013). Although some phytase products have already undergone evaluation to determine their unique P release curve, other newer products have to be thoroughly tested as they become available (Jones et al., 2010; Goncalves et al., 2016; Gourley et al., 2018; Wensley et al., 2020). GraINzyme Phytase (Agrivida, Woburn, MA) is expressed in corn that contains an engineered Escherichia coli phytase called Phy02 (Ligon, 2016). Previous research conducted by Nyannor et al. (2007) using a corn-expressed phytase (660 FTU/g) in young pig diets observed a linear increase in average daily gain (ADG) and gain-to-feed (G:F), with increasing phytase added to a diet deficient in P.

GraINzyme corn-expressed phytase improves ADG, feed efficiency, bone mineralization, and bone strength when fed to young pigs (Lee et al., 2017; Munoz Alfonso et al., 2018; Blavi et al., 2019; Broomhead et al., 2019). Previous research evaluated the effects of corn-expressed phytase on growth performance, bone characteristics, and digestibility, with other microbial phytase sources. However, these studies have only evaluated the effects of increasing dosage of GraINzyme in P-deficient diets without comparison to an inorganic P source to develop a P release curve. Therefore, the objective of this study was to evaluate the effects of GraINzyme phytase on nursery pig growth performance and bone ash to develop an available phosphorus (aP) release curve.

MATERIALS AND METHODS

The Kansas State University Animal Care and Use Committee approved the protocol (4035) used in this experiment. Corn, soybean meal, limestone, and monocalcium phosphate were analyzed for Ca and P (AOAC 985.01 and 985.01, respectively; AOAC, 2006) to determine nutrient loading values used for formulation prior to the manufacturing of diets (Table 1). Phytase was acquired for the experiment and analyzed (Agrivida Inc., Woburn, MA) using an adjusted extraction procedure that employed an optimized buffer and extraction process specifically for GraINzyme phytase (Li and Raab, 2016). Phytase activity was 9,738,000 FTU/kg. All diets were formulated to contain a Ca:P ratio of 1.10:1 with no allowance for release of Ca by phytase. The diets were corn-soybean meal-based and calculated to contain approximately 0.26% phytate P (NRC, 2012). Diets were formulated to 1.24% standardized ileal digestibility Lys with other amino acids set to meet or exceed NRC (2012) requirement estimates.

Table 1.

Analyzed ingredient composition (as-fed basis)

Ingredient	Ca, %	P, %
Limestone1	38.03	0.10
Monocalcium P1	16.73	21.73
Corn2	0.03	0.26
Soybean meal2	0.38	0.68

¹Ingredient samples were pooled and analysis was performed by the Kansas State University Soils Lab, Manhattan, KS. Values represent the mean of three samples analyzed in duplicate.

²Ingredient samples were analyzed by Hubbard Feeds, Beloit, KS.

Diet Manufacturing

GraINzyme Phytase was ground in a roller mill (California Pellet Mills, Waterloo, IA) to a similar particle size (~600 µm) as the other corn in the diet. At the time of manufacturing, five identical 1,814 kg batches of basal diet were produced and packaged in 22.7-kg bags at Hubbard Feeds in Beloit, KS (Table 2). For each experimental diet, a subset of bags from each batch of basal diet was added to the mixer along with treatment specific ingredients including sand, limestone, monocalcium P, and GraINzyme phytase to achieve the final experimental diets (Table 3). These diets were mixed at the O.H. Kruse Feed Technology Innovation Center in Manhattan, KS. Complete diet samples were taken during the bagging of experimental diets with a subsample collected from every fourth bag and pooled into one homogenized sample per dietary treatment. After homogenization, samples were stored at −20 °C until they were submitted for phytase and nutrient analysis.

Table 2.

Composition of basal mix (as-fed basis)¹

Item
Ingredient, %
Corn	64.35
Soybean meal	34.29
Sodium chloride	0.61
l-Lysine-HCl	0.31
dl-Methionine	0.12
l-Threonine	0.12
l-Valine	0.01
Trace mineral premix2	0.15
Vitamin premix3	0.05
Total	100
Calculated analysis
SID4 amino acids
Lysine, %	1.24
Isoleucine:lysine	65
Leucine:lysine	131
Methionine:lysine	34
Methionine and cysteine:lysine	58
Threonine:lysine	64
Tryptophan:lysine	19.1
Valine:lysine	71
Histodine:lysine	42
Total lysine, %	1.42
Metabolizable energy, kcal/kg	3,338
NE,5 kcal/kg	2,454
SID lysine:NE, g/Mcal	5.05
Crude protein, %	22.1
Ca, %	0.18
P, %	0.40
Available P, %	0.08
STTD P,6 %	0.17

¹The basal batch was used as the major ingredient in each experimental diet.

²Provided per kg of diet: 110 mg Zn from zinc sulfate; 110 mg Fe from iron sulfate; 33 mg Mn from manganese oxide; 17 mg Cu from copper sulfate; 0.30 mg I from calcium iodate; 0.30 mg Se from sodium selenite.

³Provided per kg of diet: 4,134 IU vitamin A; 1,653 IU vitamin D; 44 IU vitamin E; 3 mg vitamin K; 0.03 mg vitamin B12; 50 mg niacin; 28 mg pantothenic acid; 8 mg riboflavin.

⁴SID = standardized ileal digestible.

⁵NE = net energy.

⁶STTD P = Standardized total tract digestible phosphorus.

Table 3.

Ingredient composition of experimental diets (as-fed basis)¹

	Experimental diet
	Inorganic P			Phytase2
Ingredient, %	0.11	0.19	0.27	150	250	500	1,000	1,500
Basal mix	98.18	98.18	98.18	98.18	98.18	98.18	98.18	98.18
Limestone	0.69	0.77	0.83	0.69	0.69	0.69	0.69	0.69
Monocalcium P	0.16	0.53	0.90	0.16	0.16	0.16	0.16	0.16
Sand3	0.98	0.54	0.10	0.97	0.97	0.97	0.96	0.96
Phytase4	—	—	—	0.0015	0.0026	0.0051	0.0103	0.0154
Total	100	100	100	100	100	100	100	100
Calculated analysis
Crude protein, %	21.7	21.7	21.7	21.7	21.7	21.7	21.7	21.7
Ca, %	0.47	0.60	0.65	0.47	0.47	0.47	0.47	0.47
P, %	0.43	0.51	0.59	0.43	0.43	0.43	0.43	0.43
Phytase, FTU/kg	—	—	—	150	250	500	1,000	1,500
Ca:P ratio	1.10	1.10	1.10	1.10	1.10	1.10	1.10	1.10
Analyzed composition5
Ca, %	0.57	0.63	0.77	0.51	0.53	0.59	0.54	0.54
P, %	0.42	0.47	0.63	0.40	0.43	0.43	0.42	0.41
Phytase, FTU/kg6	—	—	—	137	223	379	919	1,623

¹Diets were fed for 21 d starting at approximately 10 kg.

²GraINzyme (Agrivida Inc., Woburn, MA).

³Sand was used to equalize hand-add batch including the addition of limestone, monocalcium P, and phytase when blended with the basal mix.

⁴Phytase was analyzed and contained 9,738,000 FTU/kg (Agrivida Inc., Woburn, MA).

⁵Complete diet samples were taken during bagging of experimental diets from every fourth bag and pooled into one homogenized sample per dietary treatment. Samples were stored at -20 °C until they were submitted for duplicate analysis of Ca and P (Kansas State University Soils Lab, Manhattan, KS).

⁶Five samples of each diet was submitted to Agrivida Inc. (Woburn, MA) for complete phytase analysis using an adjusted extraction procedure that employed an optimized buffer and extraction process.

Animals and Housing

The experiment was conducted at the Kansas State University Segregated Early Wean Facility in Manhattan, KS. The nursery barns were environmentally controlled, and each pen contained a four-hole, dry self-feeder, and nipple waterer for ad libitum access to feed and water.

A total of 360 barrows (Line 200 × 400, DNA, Columbus, NE; initially 9.9 ± 0.19 kg) were used in a 21-d growth trial. Pigs were weaned at approximately 21 d of age, randomly allotted to pens based on initial body weight (BW), and fed common phase 1 and 2 diets after placement for 18 d. From days 18 to 21 postweaning, all pigs were fed a diet containing 0.11% aP. On day 21 postweaning, considered day 0 of the study, pens of pigs were blocked by weight and randomly allotted to one of eight dietary treatments with five pigs per pen and nine pens (weight blocks) per treatment. Treatments consisted of three diets with increasing P from monocalcium P formulated to provide 0.11%, 0.19%, or 0.27% aP, or five diets with 150, 250, 500, 1,000, or 1,500 FTU/kg phytase added to the diet containing 0.11% aP.

During the experiment, pigs and feeders were individually weighed every 7 d to determine ADG, average daily feed intake (ADFI), and G:F. At the conclusion of the study, one pig in each pen closest to the mean BW was euthanized via penetrating captive bolt and the right fibula was collected to determine bone ash weight and percentage bone ash (72 barrows, nine observations per treatment). After collection, bones were individually placed in plastic bags with permanent identification and stored at−20 °C until processing. On the day of processing, bones were autoclaved for 1 h at 121 °C. After cooling, any leftover extraneous soft tissue including cartilage caps was cleaned from the fibulas. Fibulas (one from each pig) were dried at 105 °C for 7 d in a drying oven, reweighed, and then ashed at 600 °C for 24 h to determine total ash weight and percentage ash relative to dried bone weight (Wensley et al., 2020).

Chemical Analysis

Three samples per dietary treatment from the pooled feed samples were sent to the KSU Soils Lab in Manhattan, KS, for duplicate analysis of Ca (AOAC 985.01, 2006) and P (AOAC 985.01, 2006). Additionally, five samples of each diet were submitted for phytase analysis (Agrivida Inc., Woburn, MA) using an extraction procedure that employed an optimized buffer, (pH 10.8) as described by Li and Raab (2016), which is different from the AOAC method (Gizzi et al., 2008).

Statistical Analysis

Studentized residuals were evaluated for pen means or individual bone ash measurements to ensure data met the assumption of normal distribution. Data were analyzed as a randomized complete block design with pen as the experimental unit. An initial base model was evaluated using the MIXED procedure of SAS OnDemand for Academics (SAS Institute, Inc., Cary, NC). Treatment was considered a fixed effect and weight block a random effect. Linear and quadratic contrasts were evaluated within increasing inorganic P or phytase treatments. Contrast coefficients were adjusted to account for unequal spacing in phytase doses using the IML procedure. For pens of pigs fed the inorganic P diets, the marginal intake of aP per day was calculated for each pen using the equation: aP intake/day = [dietary aP% − 0.11% (aP in the basal diet)] × ADFI. A standard curve was then developed for each response criteria using the marginal aP release as the predictor variable. The equation for the standard curve was used to calculate aP release from each pen fed the different phytase dosages based on the observed value for each response criteria. This value was then converted to a marginal aP% using the pen ADFI.

A mixed model ANOVA with weight block as a random effect was performed to evaluate aP release as a function of phytase dosage, assuming an intercept of no aP release for the control diet without phytase. All release values were calculated using formulated P and phytase levels. The GLIMMIX procedure of SAS OnDemand for Academics (SAS Institute, Inc., Cary, NC) was used for the analysis.

Release values were used in a non-linear regression to fit a model predicting aP release curves dependent on phytase dosage as a continuous variable using the individual pen data using the following functional form:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{v}\mathrm{a}\mathrm{i}\mathrm{l}\mathrm{a}\mathrm{b}\mathrm{l}\mathrm{e}\mathrm{P}\mathrm{r}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{s}\mathrm{e},\mathrm{\%}=\frac{a\times \mathrm{F}\mathrm{T}\mathrm{U}}{b+\mathrm{F}\mathrm{T}\mathrm{U}}}\end{array}}$$

where coefficient a is a horizontal asymptote representing the maximum release of aP for the given response, and coefficient b represents the vertical asymptote. Separate aP release curves were developed using aP release derived from ADG, G:F, percentage bone ash, and bone ash weight. Model parameters were estimated using the nls function from the R Stats Package (Version 4.0.2 (2020-06-22); R Core Team, 2020) using the RStudio environment (Version 1.3.1093; RStudio Team, 2020). Results were considered significant with P-values ≤ 0.05 and considered marginally significant with P-values > 0.05 and ≤ 0.10.

RESULTS

Analysis of manufactured diets resulted in similar Ca and P values to those expected from diet formulation (Table 3). Phytase analysis of complete diets showed a stepwise increase in phytase activity as expected and were similar to formulated values for each phytase containing diet.

From days 0 to 21, pigs fed increasing aP from inorganic P had increased (linear, P ≤ 0.002) ADG, ADFI, G:F, and final BW (Table 4). In addition, pigs fed diets with increasing phytase had improved (linear, P < 0.001) growth performance across all response variables. There was a quadratic (P = 0.049) increase in G:F when the level of dietary phytase increased. For bone characteristics, bone ash weight and percentage bone ash increased (linear, P ≤ 0.003) for pigs fed either increasing inorganic P or phytase in the diet.

Table 4.

Effects of increasing aP from inorganic P or GraINzyme phytase on nursery pig growth performance and bone ash values¹,²

	Inorganic P, % aP3			Phytase, FTU/kg4						Inorganic P, P-value		Phytase, P-value
Item	0.11	0.19	0.27	150	250	500	1,000	1,500	SEM	Linear	Quadratic	Linear	Quadratic
BW, kg
Day 0	10.1	10.1	9.8	9.7	9.9	10.1	9.7	10.1	0.19	0.087	0.257	0.229	0.021
Day 21	18.7	19.8	20.4	18.9	19.0	19.5	20.0	20.5	0.36	<0.001	0.334	<0.001	0.561
Days 0 to 21
ADG, g	412	465	504	436	434	448	489	491	10.7	<0.001	0.527	<0.001	0.103
ADFI, g	686	739	746	707	717	719	739	757	15.9	0.002	0.151	0.001	0.514
G:F, g/kg	600	630	675	617	606	622	661	649	7.5	<0.001	0.404	<0.001	0.049
Bone characteristics5
Bone ash, g	0.614	0.764	0.812	0.601	0.623	0.672	0.722	0.861	0.0284	<0.001	0.141	<0.001	0.162
Bone ash, %	39.7	43.8	45.4	39.0	41.7	43.0	41.3	44.2	0.01	0.001	0.364	0.003	0.665

¹A total of 360 nursery pigs (Line 200 × 400, DNA, Columbus, NE; initially 9.9 kg body weight (BW)) were used in a 21-d growth trial to determine the available P (aP) release curve for GraINzyme phytase. There were five pigs per pen and nine replications per treatment.

²ADG = average daily gain; ADFI = average daily feed intake; G:F = gain-to-feed ratio.

³Inorganic P was added to the diet by increasing monocalcium P.

⁴GraINzyme, Agrivida, Woburn, MA.

⁵One pig per pen (nine pens per treatment) was euthanized on day 21, and the right fibula was collected to determine bone ash weight and percentage bone ash. Fibulas were autoclaved for 1 hr. After cleaning, bones were placed in a drying oven at 105 °C for 7 d and then ashed in a muffle furnace at 600 °C for 24 h.

While a given amount of GraINzyme liberates a certain amount of phytate-derived phosphorus, estimates of the relationship between that phosphorus and the apparent aP varies depending on which response criterion is measured. As the amount of phytase in the diet increased, the calculated aP release increased (linear, P ≤ 0.008) for ADG, bone ash weight, and percentage bone ash (Table 5). For G:F, there was a quadratic (P < 0.045) increase in aP release as the amount of phytase in the diet increased. At the low levels of phytase inclusion, release of aP was not evident based on negative release values for bone ash weight and bone ash percentage. However, when using the modeling approach there is evidence of a small amount of aP release at the low phytase levels for bone mineralization measures. The release equations generated from this experiment for GraINzyme for ADG (Figure 1), G:F (Figure 2), bone ash weight (Figure 3), and percentage bone ash (Figure 4) are: aP = (0.255 × FTU)/(1,299.969 + FTU), aP = (0.233 × FTU)/(1,236.428 + FTU), aP = (45,999.949 × FTU)/(462,529,200 + FTU), and aP = (0.272 × FTU)/(2,576.581 + FTU), respectively. When viewed together, the aP release for ADG and G:F is greater at any given FTU/kg compared to bone ash percentage (Figure 5). Release of aP as indicated by bone ash weight was less than ADG and G:F at lower FTU/kg levels, but increased as FTU/kg level increased relative to growth performance measures.

Table 5.

Calculated aP release values based on different response criteria¹,²

	Phytase, FTU/kg3						Probability, P-value
Item	150	250	500	1,000	1,500	SEM4	Linear	Quadratic
Performance
ADG	0.039	0.035	0.058	0.128	0.129	0.0187	<0.001	0.078
G:F	0.044	0.018	0.052	0.138	0.110	0.0163	<0.001	0.045
Bone characteristics4
Bone ash, g	−0.026	−0.008	0.032	0.074	0.182	0.0195	<0.001	0.138
Bone ash, %	−0.029	0.044	0.083	0.035	0.116	0.0339	0.008	0.721

¹The marginal intake of available P (aP) per day was calculated for each pen using the equation: aP intake/day = [dietary aP% - 0.11% (aP in the basal diet)] × ADFI. A standard curve was then developed for each response criterion using the marginal aP release as the predictor variable. The equation for the standard curve was used to calculate aP release from each pen fed the different phytase dosages based on the observed value for each response criterion.

²ADG = average daily gain. G:F = gain-to-feed ratio.

³GraINzyme, Agrivida, Woburn, MA.

⁴One pig per pen (nine pens per treatment) was euthanized on day 21, and the right fibula was collected to determine bone ash weight and percentage bone ash. Fibulas were autoclaved for 1 h. After cleaning, bones were placed in a drying oven at 105 °C for 7 d and then ashed in a muffle furnace at 600 °C for 24 h.

Figure 1.

Available P release curve for GraINzyme phytase as predicted by average daily gain.

Figure 2.

Available P release curve for GraINzyme phytase as predicted by gain-to-feed.

Figure 3.

Available P release curve for GraINzyme phytase as predicted by bone ash weight (g).

Figure 4.

Available P release curve for GraINzyme phytase as predicted by percentage bone ash.

Figure 5.

Available P release curve for GraINzyme phytase as predicted by average daily gain, gain-to-feed, percentage bone ash, and bone ash weight (g).

DISCUSSION

Swine and poultry produce low amounts of endogenous phytase and, as a result, are unable to cleave much of the phytate-bound P in cereal grains (Humer et al., 2015). Consequently, microbial phytase is commonly added to swine diets to make phytate-bound P available. Previous research has observed that there are several beneficial effects when microbial phytase is added to P-deficient diets (Dersjant-Li et al., 2015), including improved digestibility and growth performance as well as less P in swine waste, which benefits the environment (Selle and Ravindran, 2007).

The first sources of phytase were fungal-derived and commercialized in 1991 (Dersjant-Li et al., 2015). Later, Rodriguez et al. (1999) observed that E. coli phytases were more effective than fungal phytases leading to the development of new phytase sources. Phytase sources can be categorized as 3- and 6-phytases, depending on the carbon site of hydrolysis (Augspurger et al., 2003). The GraINzyme phytase used in this study, is expressed in corn, E. coli-derived, and classified as a 6-phytase.

Microbial phytase may be expressed in plants such as soybeans and canola seed. Previous research observed that phytases expressed in soybeans (Denbow et al., 1998) and canola seeds (Zhang et al., 2000) had similar performance compared to microbial phytases. However, because of the insufficient thermotolerance of these phytases, they were unable to survive the postharvest heating that occurs during the toasting of defatted meals (Haefner et al., 2005). Conversely, corn-expressed phytase, like GraINzyme, is likely to maintain efficacy after harvest because of the cooler temperature used in drying (if any) corn compared that used in the toasting of oilseed meal (Nyannor et al., 2007).

GraINzyme contains an engineered E. coli phytase called Phy02 (Ligon, 2016). Phy02 phytase is expressed in the corn (Zea mays) kernel and then ground into a course meal to improve the digestibility of phytate-bound P. Wang and Kim (2020) evaluated the growth performance and bone mineralization of poultry fed 500 to 4,500 FTU/kg of GraINzyme. Broiler chicks fed increasing levels of phytase had a linear (P < 0.001) increase in body weight gain, feed intake, and percentage bone ash.

GraINzyme has been demonstrated to improve ADG, feed efficiency, bone mineralization, and bone strength when young pigs were fed reduced P and Ca diets (Lee et al., 2017; Munoz Alfonso et al., 2018; Blavi et al., 2019). Broomhead et al. (2019) evaluated 500 to 4,000 FTU/kg from GraINzyme in a reduced P nursery pig diet. As phytase dose increased, the apparent total tract digestibility (ATTD) of P, ADG, ADFI, and percentage bone ash increased in a quadratic (P ≤ 0.003) fashion. Blavi et al. (2019) evaluated the growth performance, bone ash measurements, and digestibility of Ca and P in young pigs fed 250 to 1,500 FTU/kg GraINzyme in a Ca- and P-deficient diet. Like Broomhead et al. (2019), as phytase dose increased, the growth performance, percentage bone ash, and ATTD of Ca and P improved in a quadratic (P ≤ 0.054) fashion. Similarly, in the present study increasing phytase inclusion from 150 to 1,500 FTU/kg resulted in linear (P < 0.001) improvements in ADG, ADFI, G:F, and percentage bone ash. Similar to Broomhead et al. (2019), while there was a strong linear (P < 0.001) effect on percentage bone ash, there was a slight decrease in the pigs fed 1,000 FTU/kg of GraINzyme.

Because of the intrinsic differences between phytase sources, it is important to evaluate their unique aP release when new or enhanced phytase sources enter the marketplace. A common approach used for comparing phytase sources is to compare the phytase activity needed to reach a particular aP release value allowing products to be compared on the same level of activity. Wensley et al. (2020) observed that Smizyme TS G5 2,500 was able to achieve a 0.12 aP at 500 FTU/kg using percentage bone ash. Gourley et al. (2018) observed that Natuphos E 5,000 G was able to achieve a 0.11 aP release based on percentage bone ash with 500 FTU/kg. In our study, GraINzyme, using percentage bone ash as the criteria, would need to be added to diets at a level greater than 1,500 FTU/kg in order to achieve a similar aP release of 0.12. Using a modeling approach, the aP release increases as FTU/kg increases for all response criteria as would be expected. This is consistent with previous works with low levels of release at the lower levels and increasing release as FTU/kg increases. Depending on the response criteria used, it appears that GraINzyme may need to be included at a greater number of FTU/kg in order to obtain a similar response observed with other commercial phytase sources. One speculation is that the specific extraction method for GraINzyme may lead to higher FTU values. Based on the FTU reported by Broomhead et al. (2019), the values obtained from the Li and Raab (2016) extraction method are on average 64% greater than the values from the AOAC method. This might suggest that if the AOAC method was used, it would have likely provided lower FTU values, and that fewer FTUs are required for the same aP release.

In conclusion, using the diet formulations in the present study, an aP release curve can be developed and used for GraINzyme phytase as a means to improve aP digestibility in nursery diets when included at concentrations between 150 and 1,500 FTU/kg. While a given amount of GraINzyme liberates a certain amount of phytate-derived phosphorus, estimates of the relationship between that phosphorus and the apparent aP varies depending on which response criterion is measured.