Effect of limestone solubility on mineral digestibility and bone ash in nursery pigs fed diets containing graded level of inorganic phosphorus or increasing dose of a novel consensus bacterial 6-phytase variant

Abstract

The effect of a novel consensus bacterial 6-phytase variant (PhyG) on total tract digestibility (ATTD) of minerals and bone ash was evaluated in pigs fed diets containing medium- and high-solubility limestone (MSL and HSL, 69.6 and 91.7% solubility, respectively, at 5 min, pH 3.0) in a randomized complete block design. For each limestone, eight diets were formulated: an inorganic phosphate-free negative control (NC) based on wheat, corn, soybean-meal, canola-meal and rice-barn [0.18% standardized total tract digestible (STTD) P and 0.59% Ca]; the NC supplemented with 250, 500, 1,000, or 2,000 FTU/kg of PhyG, and; the NC with added monocalcium phosphate (MCP) and limestone to produce three positive controls (0.33, 0.27, and 0.21% STTD P, and 0.75, 0.70, and 0.64% Ca, respectively; PC1, PC2, PC3). In total, 128 pigs (12.8 ± 1.33 kg, 8 pigs/treatment, housed individually) were adapted for 16 d followed by 4 d of fecal collection. Femurs were collected from euthanized pigs on day 21. Data were analyzed by one-way ANOVA with means separation by Tukey’s test, and by factorial analysis (2 x 4: 2 levels of limestone solubility, 4 STTD P levels, and 2 × 5: 2 levels of limestone solubility, 5 PhyG dose levels). Phytase dose-response was analyzed by curve fitting. A consistent negative effect of HSL on ATTD P and Ca was observed in control diets (P < 0.001). Across phytase-supplemented diets, HSL reduced (P < 0.05) ATTD Ca and P (% and g/kg) compared with MSL. Across limestones, increasing phytase dose level increased (P < 0.05) ATTD P exponentially. Limestone solubility had no effect on bone ash, but PhyG linearly increased (P < 0.05) bone ash; 500 FTU/kg or higher maintained bone ash (g/femur) equivalent to PC1. In conclusion, ATTD P and Ca were reduced by a high compared with a medium soluble limestone, but the novel phytase improved ATTD P and Ca independent of limestone solubility.

This study has generated the first data in young pigs showing an effect of limestone solubility on mineral digestibility. A highly soluble limestone reduced phosphorus and calcium digestibility compared to a less soluble limestone, however this effect was reduced by increasing the dose level of the phytase.

Introduction

Calcium (Ca) and phosphorus (P) are essential for healthy bone and soft tissue development (González-Vega and Stein, 2014). In pigs, up to 99% of Ca and 80% of P is stored in bone (Crenshaw, 2001). Cereals and oilseeds that form the basis of commercial pig diets are inherently low in Ca. Hence, Ca is commonly added to the diet to ensure animal requirements are met. Usually, this is in the form of limestone (CaCO₃). On the other hand, the P content of cereals and oilseeds is high, but up to 80% is locked up in the form of phytate (myo-inositol hexakisphosphate, IP₆) which is poorly digested by monogastric animals (Selle and Ravindran, 2007). Inorganic P is also usually added to the diet to supplement requirements but is costly.

Digestion and metabolism of Ca and P are complex and closely interrelated (Li et al., 2017). An excess or deficiency in either one can lead to compromised utilization of the other (Létourneau-Montminy et al., 2015). Excess Ca binds to free phosphate (Hurwitz and Bar, 1971), reducing P-availability for absorption. It also binds to phytate, forming Ca-phytate complexes which are insoluble at or above pH 5 and therefore hard to digest (Selle et al., 2009). High dietary Ca, or imbalance with P, reduces the bioavailability of both Ca and phytate-derived P (Lei et al., 1994; Liu et al., 1998). Formulating diets with these minerals in balance is therefore key for optimal nutrition, especially when dietary P is close to requirement levels (Qian et al., 1996).

The impact of limestone quality on mineral digestibility has been little researched in pigs. Limestone sources vary in their composition, purity and particle size, the latter being inversely correlated to solubility (Zhang and Coon, 1997; Walk et al, 2012; Kim et al., 2018). These differences have the potential to affect bioavailability of the Ca and its interactions with other nutrients and so to affect mineral availability and utilization. In the literature, limestones are variously described as ‘fine’, ‘medium’, or ‘coarse’. ‘Fine’ limestones (typically with particle size below < 150 µm in diameter) are favored because they are less likely to damage feed processing machinery than ‘coarse’ limestones (typically with particle size of 300 to 500 µm in diameter). However, in boilers, in vitro and in vivo studies have clearly demonstrated that smaller limestone particle sizes with higher solubility can reduce ileal digestibility (AID) and retention of Ca and P as well as weight gain and bone ash compared to coarser limestones with lower solubility (Manangi and Coon, 2007; Anwar et al., 2016, 2017; Kim et al., 2018). A similar effect in pigs is unproven but plausible given the known influence of dietary Ca concentration on mineral digestibility in both species.

Understanding the influence of limestone solubility on nutrient digestibility and utilization is particularly relevant in the context of microbial phytase supplementation because phytase is so widely used to increase P-availability in modern pig diets. Limited evidence from poultry studies suggests that phytase efficacy is affected by limestone particle size and/or solubility (Kim et al., 2018; Davin et al., 2020; Majeed et al., 2020) but no comparable data are available for pigs.

In the current study, the effect of limestone solubility on total tract digestibility (ATTD) of P and Ca and bone ash content in nursery pigs was evaluated in the presence of a novel consensus bacterial 6-phytase variant (PhyG), applied at 5 dose levels in diets containing no added inorganic phosphate (Pi). The primary objective was to determine whether and how Ca and P digestibility are affected by dietary inclusion of a limestone with high solubility and small particle size compared to one with medium solubility and larger particle size, in the presence of the phytase at increasing dose levels between 0 and 2,000 phytase units (FTU)/kg. Three positive control diets containing increasing levels of Ca and standardized total tract digestible (STTD) P were included as a reference and used for estimation of the digestible P equivalence values of the phytase at different dose levels, as a secondary objective.

Materials and Methods

The experimental protocols used in the study were approved by the Animal Experiments Committee of The University of Queensland, Australia. Pigs were cared for according to the Animal Care and Protection Act (2001) and the Australian Code for the Care and Use of Animals for Scientific Purposes (2013).

Pigs, housing and experimental design

A total of 128 Large White pigs (initial BW 12.8 ± 1.33 kg) were randomly assigned to 16 dietary treatments with eight replicate pigs per treatment in a randomized complete block design. Body weight was the blocking factor. Pigs were housed in individual floor pens in an environmentally controlled animal rearing house in which temperature was maintained at 24 to 27 °C. The lighting regime was LD 16:8 h. Pens had slatted floors. Water and feed were provided ad libitum for the duration of the experiment (20 d). Water was accessed via a nipple drinker system.

Dietary treatments

The 16 dietary treatments comprised 8 diets containing a high-solubility limestone (HSL) and 8 diets containing a medium solubility limestone (MSL). The HSL had an average particle size of 56.1 μm (geometric mean diameter, GMD) and was sourced from Warwick mine (Sibelco Warwick, Australia), whereas the MSL had an average particle size of 91.8 μm (GMD) and was sourced from Riverton mine (Sibelco Australia, Australia). In each case, the eight treatment diets comprised of: a negative control (NC) based on wheat, corn, soybean-meal, canola-meal, and rice bran, containing no added Pi and reduced in Ca (0.18% STTD P, 0.59% Ca) versus the positive control (PC) diets; the NC supplemented with 250, 500, 1,000, or 2,000 FTU/kg of a novel consensus bacterial 6-phytase variant produced in Trichoderma reesei, PhyG (Danisco Animal Nutrition & Health, IFF Inc., New York), and; the NC with added MCP and limestone at three levels to produce three positive control diets (PC1, PC2, PC3) containing 0.33, 0.27, and 0.21% STTD P, and 0.75, 0.70, and 0.64% Ca, respectively. All diets contained titanium dioxide added at 0.5% and celite (a source of acid-insoluble ash) added at 2.0%, as indigestible markers. The composition of the treatment diets is given in Table 1.

Table 1.

Ingredient composition (%, as-fed basis), and calculated nutrient content (%) of the basal diets.

	PC1 0.33% STTD1 P	PC2 0.27% STTD1 P	PC3 0.21% STTD1 P	NC 0.18% STTD1 P
Ingredient, %
Corn	10.00	10.00	10.00	10.00
Wheat	51.01	51.01	51.01	51.01
Soybean meal, 46% CP	20.37	20.37	20.37	20.37
Canola-meal	8.00	8.00	8.00	8.00
Soy oil	2.75	2.75	2.75	2.75
Rice bran	2.00	2.00	2.00	2.00
Limestone2	1.44	1.42	1.40	1.30
Monocalcium phosphate	0.770	0.452	0.134	-
Bentonite	-	0.427	0.851	1.03
Salt	0.200	0.200	0.200	0.200
L-Lysine HCL	0.382	0.382	0.382	0.382
DL-Methionine	0.077	0.077	0.077	0.077
Threonine	0.142	0.142	0.142	0.142
Biotronic acidifier	0.100	0.100	0.100	0.100
Elitox-toxin binder	0.050	0.050	0.050	0.050
Vitamin-mineral premix3	0.200	0.200	0.200	0.200
Choline chloride	0.010	0.010	0.010	0.010
Titanium dioxide	0.500	0.500	0.500	0.500
Celite	2.000	2.000	2.000	2.000
Calculated composition, %
Dry matter	89.66	89.66	89.66	89.66
Crude protein	19.54	19.54	19.54	19.54
Crude fiber	3.46	3.46	3.46	3.46
Fat	4.49	4.49	4.49	4.49
NE, kcal/kg	2,400	2,400	2,400	2,400
SID Lysine	1.14	1.14	1.14	1.14
SID Methionine	0.341	0.341	0.341	0.341
SID Methionine +cysteine	0.663	0.663	0.663	0.663
SID Isoleucine	0.682	0.682	0.682	0.682
SID Leucine	1.26	1.26	1.26	1.26
SID Threonine	0.716	0.716	0.716	0.716
SID Tryptophan	0.203	0.203	0.203	0.203
SID Valine	0.772	0.772	0.772	0.772
SID Arginine	1.09	1.09	1.09	1.09
Calcium	0.745	0.695	0.640	0.585
Total phosphorus	0.623	0.551	0.479	0.448
Ca:total P	1.20	1.26	1.34	1.31
STTD phosphorus	0.330	0.27	0.210	0.183
Phytate-phosphorus	0.298	0.298	0.298	0.298

STTD, standardized total tract digestibility.

Medium or high solubility, depending on the treatment.

Vitamin-mineral premix composition: Active level/kg premix: Vitamin A (MIU) 5, Vitamin D (MIU) 1, Vitamin E (g) 15, Vitamin K (g) 0.75, Nicotinic Acid B3 (g) 10, Pantothenic Acid B5 (g) 6, Folic Acid (g) 0.50, Riboflaven B2 (g) 2, Cyanocobalamin B12 (g) 0.01, Biotin (g) 0.05, Pyridoxine B6 (g) 1.5, Thiamin B1 (g) 1.5, Copper (g) 10, Cobalt (g) 0.25, Ioden (g) 0.5, Selenium (g) 0.1, Iron (g) 35, Zinc (g) 45, Manganese (g) 25, Antioxidant (g) 10.

Measurements and sampling

Pigs were weighed at the beginning and at the end of the experiment. The initial 16 d was considered the adaptation period to the experimental diets. Fecal samples were collected via anal stimulation once daily on days 17, 18, 19, and 20, and stored at -20 °C for later determination of ATTD of dry matter (DM), P and Ca. On day 21, pigs were weighed, feeders emptied, and the amount of feed left in each feeder was recorded and subtracted from the total feed allotment to calculate feed disappearance in each pen. Pigs were euthanized by Pentobarbitone Sodium injection and the right femur was collected immediately and frozen at -20 °C for later defatting and determination of ash content.

Samples of the final diets were analyzed for DM, N, Ca, P, phytate-P, and phytase activity. Fecal samples were analyzed for DM, N, Ca, and P.

Chemical analysis

The in vitro solubility of the two limestone sources was determined at 5, 15, and 30 min of incubation at pH 3.0 at 42 ºC, according to the method described by Kim et al. (2019). Analyses were conducted in duplicate.

Fecal samples were dried in a forced-air oven at 50 °C. Feed and fecal samples were ground using a Wiley Mill with a 1-mm screen. All analyses were conducted at the University of Queensland, Australia, with the exception of phytase activity and phytate-P content of the diets which were analyzed by Danisco Animal Nutrition Research Centre, Brabrand, Denmark. Calcium and P in feed and fecal samples were determined by Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES) following microwave assisted acid digestion, according to AOAC method 985.01 (AOAC, 1996). Nitrogen was determined by AOAC method 990.03 (AOAC, 2006). Dry matter in feed and fecal samples was determined according to AOAC method 930.15 (AOAC, 2005). Titanium in feed and feces was determined according to the procedures described by Lomer et al. (2000) and measured on an inductively coupled plasma mass spectrometer. Phytase activities in final diets were analyzed according to a modified version of AOAC method 2000.12. (Engelen et al., 2001). One phytase unit (FTU) was defined as the quantity of enzyme that released 1 μmol of inorganic phosphate per minute from 5.0 mmol/L sodium phytate substrate at pH5.5 at 37 °C. Phytate-phosphorus concentrations in diets were determined using a modified version of the HPLC method described by Skoglund et al. (1997).

Frozen femur samples were thawed and autoclaved at 125 °C for 55 min. Femurs were broken, dried, and soaked for 72 h in petroleum ether under a chemical hood to remove marrow and fat. Femurs were then dried for 2 h at 135 ºC and ashed at 600 ºC for 16 h in a muffler furnace.

Calculations

The ATTD of nutrients was calculated based on the concentrations of the titanium dioxide marker and of the nutrient in the feed and in the fecal samples, according to the following formula:

Apparent total tract digestibility of nutrients (ATTD, %) = 100 − {[(Nd/Nf) × (Tif/Tid)] × 100},

Where Nd = nutrient concentration in the feces (mg/kg DM), Nf = nutrient concentration in feed (mg/kg DM), Tif = titanium concentration in feed (mg/kg DM), Tid = titanium concentration in the feces (mg/kg DM).

Statistical analysis

Data were analyzed on a per pig basis. Variability was expressed as a pooled SEM. Data were analyzed by one-way ANOVA and means were separated by Tukey’s HSD test. In addition, a 2 × 4 factorial analysis was performed to test the effect of limestone solubility at different MCP levels in control diets on response measures (PhyG treatments excluded), and a 2 × 5 factorial analysis was performed to determine the effect of limestone solubility on response measures at different phytase dose levels (PC treatments excluded). Phytase dose-response was analyzed by curve fitting. All data analyses were performed in the Fit Model platform of JMP 14.0 (SAS Institute Inc., Cary, NC, 2014). Linear regression of bone ash (%) against increasing STTD P from MCP in the control diets (NC, PC1, PC2, PC3) was performed as a reference for calculating the digestible P equivalence of PhyG. This was calculated by applying the response parameter (bone ash %) value at a given phytase dose level and calculation of the corresponding MCP-P replacement value. The significance of main and interactive effects was determined at P < 0.05, whereas 0.05 < P < 0.1 was considered a tendency.

Results

Limestone composition, solubility, and particle size

The analyzed composition and average particle sizes of the two limestone sources are presented in Table 2. Their solubilities after 5, 15, and 30 min at 42 °C and pH 3.0 are given in Figure 1. The limestones had comparable Ca content (38.60% and 38.62% for MSL and HSL, respectively). The MSL had a higher content of Fe, Mn, Cu and a higher moisture content (0.11% vs. 0.01%) than the HSL. The GMD of the limestone particles was 91.77 μm for MSL and 56.13 μm for HSL. The HSL dissolved more rapidly than the MSL. After incubation for 5 min, the solubilities were 69.6% and 91.7%, for MSL and HSL, respectively (Figure 1).

Table 2.

Particle size and chemical composition of the two limestone sources.

	Medium solubility limestone1	High solubility limestone2
Particle size, geometric mean diameter (GMD), μm	91.77	56.13
Dry matter, %	99.89	99.99
Moisture, %	0.11	0.01
Calcium, %	38.60	38.62
Phosphorus, %	0.02	0.01
Magnesium, %	0.15	0.22
Potassium, %	0.03	0.02
Sodium, %	0.01	0.01
Iron, mg/kg	1,078.25	179.18
Manganese, mg/kg	258.41	9.13
Zinc, mg/kg	11.96	13.95
Copper, mg/kg	2.28	0.01

69.6% soluble after 5 mins incubation at pH 3.0. Source: Riverton mine (Sibelco Australia, Australia).

91.7% soluble after 5 mins incubation at pH 3.0. Source: Warwick mine (Sibelco Warwick, Australia).

Figure 1.

In vitro solubility of two limestones sources, at 42°C, pH 3.0. High-solubility limestone: Sibelco Warwick, Australia. Geometric mean diameter (GMD) 91.77 μm; medium solubility limestone: Sibelco Australia, Riverton, Australia, GMD 56.13 μm.

Analyzed phytase and nutrients in the diets

The concentrations of analyzed nutrients in the PC and NC diets are presented in Table 3. Values of analyzed P in the NC diets and PC HSL diets were comparable with calculated values, whereas those in the PC MSL diets were slightly higher than calculated [range +0.04 to +0.07 percentage points (pp)]. Analyzed concentrations of Ca were consistently greater than calculated values (+0.10 to +0.17 pp) but were similar in the NC MSL and HSL diets (0.008 pp difference). The analyzed Ca:P ratios in the NC MSL and HSL diets were equal (1.53 in both cases) whilst those in the PC MSL diets ranged from 1.28 to 1.44 and those in the PC HSL diets ranged from 1.36 to 1.58. There was no overlap between control treatments in their analyzed P or Ca content and the concentration of both minerals decreased in a stepwise manner from PC1 to PC2, PC3 and NC, as expected. Analyzed N content was comparable across diets, and phytate-P levels were also in close agreement with calculated levels (range 0.29% to 0.31% DM; Table 3). Phytase activities in the PhyG supplemented diets were in broad agreement with target dose levels after accounting for endogenous phytase activity in the basal NC diets (range 79% to 135% of target dose levels across the phytase-supplemented diets) (Table 4).

Table 3.

Analyzed nutrient concentrations (%, DM basis) of the basal diets.

	PC1 0.33% STTD1 P	PC2 0.27% STTD1 P	PC3 0.21% STTD1 P	NC 0.18% STTD1 P
Medium solubility limestone2
Nitrogen	3.98	3.95	3.95	3.98
Calcium	0.850	0.827	0.796	0.743
Phosphorus	0.660	0.600	0.552	0.487
Phytate-P	0.300	0.310	0.310	0.310
Highly solubility limestone3
Nitrogen	4.01	3.98	3.95	3.99
Calcium	0.831	0.812	0.804	0.751
Phosphorus	0.609	0.548	0.509	0.491
Phytate-P	0.310	0.290	0.300	0.290

STTD, standardized total tract digestibility.

69.6% soluble after 5 mins incubation at pH 3.0, geometric mean diameter (GMD) 91.77 μm.

91.7% soluble after 5 mins incubation at pH 3.0, GMD 56.13 μm.

Table 4.

Analyzed phytase activities in the treatment diets.

Diets	Limestone solubility after 5 min1, %	STTD P, %	PhyG, FTU/kg	Analyzed phytase activity, FTU/kg
PC1	69.6	0.33	0	489
PC2	69.6	0.27	0	416
PC3	69.6	0.21	0	403
NC	69.6	0.18	0	347
NC	69.6	0.18	250	617
NC	69.6	0.18	500	906
NC	69.6	0.18	1,000	1,494
NC	69.6	0.18	2,000	2,289
PC1	91.7	0.33	0	499
PC2	91.7	0.27	0	440
PC3	91.7	0.21	0	407
NC	91.7	0.18	0	326
NC	91.7	0.18	250	523
NC	91.7	0.18	500	1,001
NC	91.7	0.18	1,000	1,417
NC	91.7	0.18	2,000	1,958

at pH 3.0.

Total tract digestibility of minerals

Effects of treatment on ATTD of DM, P, and Ca are presented in Table 5. The 2 x 4 factorial analysis that excluded PhyG treatments confirmed an independent effect of limestone solubility on ATTD of P and Ca (P < 0.001, % or g/kg basis) in the control treatments. According to this analysis, MSL increased ATTD of P and Ca compared with HSL by 5.4 and 3.6 pp or 0.46 and 0.35 g/kg (P < 0.05, Table 5), respectively. Varying the STTD of P (%) in the diet also affected ATTD of P (% or g/kg; P < 0.01) and Ca (g/kg; P < 0.01); reduced dietary STTD of P within the range 0.27% to 0.18% resulted in a reduction in the digestible content of both minerals.

Table 5.

Apparent total tract digestibility (ATTD) of calcium and phosphorus in pigs fed PhyG supplemented diets containing medium- and high-solubility limestone, at 55 d of age

Diets	Limestone solubility1 after 5 min, %	STTD P, %	PhyG, FTU/kg	DM, %	ATTD P, %	DigestibleP, g/kg	ATTD Ca, %	Digestible Ca, g/kg
PC1	69.6	0.33	0	84.33a	50.81cd	3.35a	49.97bcdef	4.42a
PC2	69.6	0.27	0	84.15a	50.17 cd	3.01abc	49.38cdef	4.08abc
PC3	69.6	0.21	0	83.17ab	47.75de	2.64cd	50.29bcdef	4.00abc
NC	69.6	0.18	0	82.79abcd	40.74ef	1.98e	47.99ef	3.57bcd
NC	69.6	0.18	250	82.99abc	55.90bc	2.72bcd	57.88ab	4.30a
NC	69.6	0.18	500	83.10ab	61.39ab	2.99abc	56.29abcd	4.18a
NC	69.6	0.18	1,000	82.42bcde	63.94a	3.11ab	56.83abc	4.22a
NC	69.6	0.18	2,000	82.57bcde	67.67a	3.30a	58.47a	4.35a
PC1	91.7	0.33	0	83.46ab	44.10def	2.69cd	47.73ef	3.97abc
PC2	91.7	0.27	0	84.31a	45.96def	2.52d	48.85def	3.97abc
PC3	91.7	0.21	0	82.10bcde	39.16f	1.99e	43.46ef	3.50cd
NC	91.7	0.18	0	81.45cde	38.83f	1.91e	43.05f	3.23d
NC	91.7	0.18	250	81.17e	50.53cd	2.48d	51.32abcde	3.85abc
NC	91.7	0.18	500	82.84abcd	61.55ab	3.02abc	55.41abcde	4.16ab
NC	91.7	0.18	1,000	81.32de	61.71ab	3.03abc	53.43abcde	4.01abc
NC	91.7	0.18	2,000	81.91bcde	63.47a	3.12ab	53.12abcde	3.99abc
SEM				0.318	1.518	0.081	1.568	0.123
P-value, ANOVA				<0.001	<0.001	<0.001	<0.001	<0.001
2 x 4 factorial analysis (PhyG treatments excluded)
Main effect means
STTD of P
PC1		0.33		83.90a	47.46a	3.01a	48.85	4.19a
PC2		0.27		84.23a	48.07a	2.76a	49.11	4.02ab
PC3		0.21		82.64b	43.46ab	2.31b	46.88	3.75bc
NC		0.18		82.12b	38.79b	1.95 c	45.52	3.40c
Limestone solubility	69.6			83.61a	47.37a	2.74a	49.41a	4.02a
	91.7			82.83b	42.01b	2.28b	45.77b	3.67b
P-value, dietary STTD P level				<0.001	<0.001	<0.001	0.118	<0.001
P-value, limestone solubility				0.003	<0.001	<0.001	0.004	0.006
P-value, dietary STTD P level x limestone solubility				0.165	0.192	0.006	0.257	0.501
2 x 5 factorial analysis (PC treatments excluded) excluded)
Main effect means
Phytase dose level			0	82.12ab	38.79c	1.95c	45.52b	3.40b
			250	82.08ab	53.22b	2.60b	54.60a	4.08a
			500	82.97a	61.47a	3.01a	55.85a	4.17a
			1,000	81.87b	62.82a	3.07a	55.13a	4.12a
			2,000	82.24ab	65.57a	3.21a	55.79a	4.17a
Limestone solubility	69.6			82.77a	57.93a	2.82a	55.49a	4.12a
	91.7			81.74b	55.22b	2.71b	51.26b	3.85b
P-value, phytase dose level				0.031	<0.001	<0.001	<0.001	<0.001
P-value, limestone solubility				<0.001	0.005	0.018	<0.001	<0.001
P-value, phytase dose level x limestone solubility				0.083	0.381	0.381	0.342	0.336

at pH 3.0.

Means within a column lacking a common letter are significantly different (P < 0.05).

When only the NC and PhyG treatments were included in the analysis (2 x 5 factorial), a significant effect of limestone solubility on ATTD of P and Ca (% or g/kg basis) was observed (P < 0.05 in all cases); across phytase dose levels, MSL (vs. HSL) increased ATTD of P by 2.71 pp or 0.11 g/kg, and increased ATTD of Ca by 4.23 pp or 0.27 g/kg. Phytase dose levels affected ATTD of P and Ca (% and g/kg basis; P < 0.001); increasing the dose level within the range 0 to 2,000 FTU/kg increased ATTD of these minerals across MSL and HSL diets (+24.03 or + 26.78 pp for ATTD P with PhyG at 1,000 or 2,000 FTU/kg, respectively, vs. NC). No interaction between limestone solubility and phytase dose level was observed for either ATTD Ca or P (% or g/kg basis).

The effects of increasing PhyG dose level on ATTD of P and Ca in diets containing MSL separately to those containing HSL, modeled by exponential curve fitting, are presented in Table 6 and Figure 2. Significant exponential relationships were observed for both MSL and HSL-containing diets (P < 0.001; Figure 2); ATTD of P increased exponentially with increasing PhyG dose level in both MSL and HSL-containing diets, the values of ATTD P were higher with MSL. However, the growth rate of the curve was different therefore the difference on ATTD P between MSL and HSL was reduced at high phytase dose (Figure 2). No significant relationship between PhyG dose level and ATTD of Ca was evident in diets containing MSL. However, in diets containing HSL, there was a significant exponential relationship between increasing phytase dose level and ATTD of Ca in g/kg (P < 0.05; Table 6).

Table 6.

Fitted exponential response curves¹ for the relationship between analyzed phytase activity (above NC) and apparent total tract digestibility (ATTD) of Ca and P (g/kg) and bone ash (%), in pigs fed diets containing medium (MSL) or high-solubility (HSL) limestone.

Y	Parameters of the exponential fitted curve1			P value, exponential growth rate	R square value
	a	b	c
ATTD P, g/kg
MSL2	3.229	-1.235	-0.003	<0.001	0.86
HSL3	3.176	-1.289	-0.002	<0.001	0.81
ATTD Ca, g/kg
MSL2	4.301	-0.731	-0.007	0.093	0.53
HSL3	4.050	-0.821	-0.005	0.038	0.58
Bone ash, %
MSL2	47.357	-7.492	-0.001	0.021	0.63
HSL3	46.574	-7.662	-0.002	0.025	0.52

Fitted exponential model was a + b × Exp (c × analyzed phytase above NC).

69.6% soluble after 5 mins incubation at pH 3.0, geometric mean diameter (GMD) 91.8 μm.

91.7% soluble after 5 mins incubation at pH 3.0, GMD 56.1 μm.

Figure 2.

Effect of PhyG dose-level¹ on ATTD P (g/kg DM) in nursery pigs fed with diets containing medium³ or high⁴ solubility limestone, at 55 d of age.

Bone ash

There was no interaction but a significant effect of dietary treatment (data for MSL and HSL pooled) on bone ash (% and g/femur) at 55 d of age [P < 0.001, Figure 3a and b]. Decreasing the STTD P content of the control diet from 0.33% (PC1) to 0.18% (NC) resulted in stepwise decreases in bone ash (% and g/femur); bone ash of pigs fed the NC was 4.77 pp or 4.62 g/femur lower than that of pigs fed PC1 (ANOVA P < 0.05). Increasing the dose level of PhyG from 0 (NC) to 2,000 FTU/kg exponentially increased bone ash across both limestone solubility levels. Compared to PC, PhyG at or above 250 FTU/kg maintained bone ash % equivalent to PC1 whereas on a g/femur basis this was not achieved until 500 FTU/kg. At 2,000 FTU/kg, bone ash was increased beyond the level of PC1 on a percentage basis [P < 0.05, Figure 3a] or equivalent to PC1 on a g/femur basis [P > 0.05, Figure 3 b]. The relationship between PhyG dose level and bone ash (%) is modeled separately for MSL and HSL treatments in Table 6. The fitted curves showed an exponential relationship in both cases (P < 0.05) whereby increasing the dose level of PhyG was correlated with increasing bone ash percentage.

Figure 3.

Effect of dietary treatment on bone (femur) ash of nursery pigs at 55 d of age (medium- and high-solubility limestone data pooled) on a percentage (a) and g/femur (b) basis.

There was no significant effect of limestone solubility on bone ash (g/femur) either when analyzed as a 2 x 4 factorial (PhyG treatments excluded) or a 2 x 5 factorial (PhyG treatments included, Figure 4).

Figure 4.

Effect of limestone solubility (medium¹ versus high²) on bone ash (g/femur) of nursery pigs at 55 d of age.

Digestible P equivalence

The digestible inorganic P equivalence (g/kg diet) of PhyG in diets containing MSL and HSL was calculated based on bone ash (%) as the response parameter, using the observed responses to increasing STTD of P from MCP in the control diets (NC and PC3, PC2, and PC1) as a reference. This dose-response relationship was linear for both MSL and HSL treatments (Figure 5). Calculated digestible P equivalence values (g/kg) are given in Table 7. Regardless of limestone solubility, PhyG P equivalence values increased with increasing phytase dose level. At 1,000 FTU/kg, the digestible P equivalence values of PhyG when added to diets containing MSL and HSL were 1.89 and 2.32 g/kg diet, respectively.

Figure 5.

Linear regression on bone ash with increasing standardized total tract digestibility (STTD) of P from MCP above negative control diet. Linear regression was performed with increasing added STTD P from MCP against bone ash, with an equation y = a + bx, where y is the response parameter and x is the increasing added STTD P from MCP.

Table 7.

Calculated digestible P equivalence values (STTD) (g/kg diet) of PhyG in diets containing medium¹ or high² solubility limestone, based on bone ash as the response parameter, using increasing standardized total tract digestibility of P from MCP as a reference³.

PhyG dose level, FTU/kg	Medium solubility limestone1	High solubility limestone2
250	0.744	1.082
500	1.253	1.727
1,000	1.894	2.317
2,000	2.413	2.581

69.6% soluble after 5 mins incubation at pH3.0, geometric mean diameter (GMD) 91.77 μm.

91.7% soluble after 5 mins incubation at pH3.0, geometric mean diameter (GMD) 56.13 μm.

The digestible P equivalence as calculated by applying y values at a given phytase dose (analyzed dose) and calculating the corresponding x values based on the linear regression model presented in Fig. 1.

Discussion

The analyzed Ca content of the two limestone sources was similar (varying by only 0.02%), within the range of published values for commercial limestones used in animal feed (Pelicia et al., 2009; Browning and Cowieson, 2014; Davin et al., 2020) and in close agreement with the published NRC (1994) value of 380 g/kg. The two limestones originated from the same geographical region (Australia) but not the same source location. This may account for the observed differences in their micromineral content that is likely influenced by the local geology of the limestone rock. Aside from Ca as the major constituent, trace minerals including Fe and Mn were also present in measurable quantities, as previously noted by Davin et al. (2020) and Anwar et al. (2016).

Both limestone sources would be regarded as ‘fine’ in a commercial context. The difference in their mean particle sizes (91.77 μm GMD for MSL and 56.13 μm GMD for HSL) was modest relative to that of some other limestones used in the existing literature that relates to broilers (Kim et al. 2018, 2019; Davin et al., 2020; Li et al., 2021). Nevertheless, a clear difference in their in vitro solubilities was observed (+22% higher for the limestone with smaller particle size), that was consistent with the known inverse relationship between limestone particle size and in vitro solubility (Zhang and Coon, 1997; Kim et al., 2018). Given that limestone is the major source of dietary Ca in pigs, it was considered plausible that such differences in in vitro solubility may affect Ca availability and utilization in vivo. Specifically, a limestone with higher solubility was expected to lead to a greater amount of Ca in the gut being solubilized, raising the local pH and leading to increased formation of stable Ca-phytate complexes and the binding of excess Ca to free P, both of which may reduce the bioavailability of P and Ca (Selle et al., 2000).

The values of ATTD of P from the NC (without phytase) were (40.7% with MSL and 38.8% with HSL) broadly comparable with values obtained from NC diets in several other studies involving pigs of similar starting weight fed corn-soybean meal-based diets (Rosenfelder-Kuon et al., 2020a; Espinosa et al., 2021; Velayudhan et al., 2021). They are also consistent with the findings of a recent meta-analysis by Rosenfelder-Kuon et al. (2020b), in which an average ATTD P of ~40% was predicted. Values of ATTD Ca in the NC (48.0% with MSL and 43.1% with HSL) were lower than expected but also broadly comparable with the findings of Rosenfelder-Kuon et al. (2020a), Espinosa et al. (2021), and Velayudhan et al. (2021). These studies and the present study all used the indigestible marker approach (titanium dioxide) for calculating digestibility rather than the total fecal collection. It is possible that this method produced lower estimates of ATTD of Ca, although TiO₂ recovery was high (above 95%) in the current study.

In the control diets that contained no phytase, there was no consistent overarching effect (across limestones) of increasing STTD P content on ATTD of P or Ca (when expressed as %). However, within HSL or MSL treatments, digestible P (g/kg) was higher in PC1 than PC3 and digestible Ca (g/kg) tended to be higher, which is due to the increased P-availability of these diets. In addition, there was a clear difference in ATTD P (g/kg and, less obviously on % basis) between the HSL and MSL control (PC) treatments: ATTD P was higher in diets containing MSL than in the corresponding diets containing HSL. The analyzed P content of the MSL (but not HSL) PC diets was slightly higher than calculated (range +0.04 to +0.07 pp), but since analyzed Ca was also slightly above expectations this did not affect the Ca:P ratios, which were comparable across HSL and MSL PC diets (range 1.28 to 1.44 and 1.36 to 1.58, respectively). The lower ATTD P in PC diets containing HSL is consistent with the hypothesis of HSL conferring increased Ca solubilization, increased Ca-phytate complexing, and reduced availability of P from phytate by consequence. Several previous authors have observed a similar effect of limestone solubility in broilers: Kim et al. (2018) observed reduced AID P in broilers fed pulverized (< 75 μm, GMD) compared with particulate (402 μm) limestone (independent of phytase effects); Kim et al. (2019) observed an 83% reduction in AID P in broilers fed a highly ground limestone (0.063 mm GMD) compared with the unground limestone (0.633 mm GMD), and; Li et al. (2021) reported a lower standardized ileal digestibility (SID) of P with diets containing a fine (151 μm GMD) compared with a coarse (800 μm GMD) limestone. The interaction between limestone solubility and STTD P level on ATTD P (g/kg basis) in the control diets appeared to be such that the negative impact of HSL (vs. MSL) was greater at higher STTD P levels, which could be explained by higher Ca content. The comparable literature in pigs is limited to a single study: Merriman and Stein (2016) reported no influence of limestone particle size (200, 500, 700, or 1,125 μm, associated solubility characteristics were not stated) on ATTD or retention of P at d 21, although the corn-potato protein isolate semisynthetic diet used with low phytate-P content might have reduced the likelihood of the larger, more soluble, limestones from having a measurable impact on mineral digestibility. In addition, in that study the limestone particle size tested is higher than the current study. More studies are needed to establish the consistency and extent of the limestone solubility effect on mineral digestibility in pigs.

Effects of limestone solubility on Ca digestibility have also been little studied in pigs. In broilers, Anwar et al. (2016) demonstrated an inverse relationship between limestone in vitro solubility and in vivo Ca availability and digestibility (AID Ca). Several other studies have reported a similar effect (Kim et al., 2019; Davin et al., 2020; Li et al., 2021). The effect (in broilers) may also be dependent on dietary Ca concentration since Kim et al. (2018) observed reduced Ca digestibility (without phytase) in broiler diets containing pulverized versus particulate limestone when the dietary Ca content was low (0.6% Ca), but not when it was higher (0.8 or 1.0% Ca). In the present study in pigs, ATTD Ca was lower in HSL control compared with MSL treatments which provide an indication of a similar negative effect of HSL on Ca digestibility in pigs as that which has been observed in broilers.

The effect of phytase (PhyG) inclusion in the NC diets, independent of limestone solubility effects, was to increase the total tract digestibility of P in a positive dose-dependent manner that was exponential for ATTD P on a g/kg basis. The improvements in ATTD with PhyG at 1,000 or 2,000 FTU/kg compared to the NC without phytase were substantial (+24.03 and +26.78 pp, respectively, on average across MSL and HSL diets). This suggests that the phytase exhibited a high capacity to increase total tract digestibility of P in the Pi-free diets employed in the present study, as has previously been shown by Velayudhan et al. (2021) with the same phytase at 500 and 1,000 FTU/kg in grower–finisher pigs. A recent meta-analysis of 88 studies incorporating a variety of different dietary presentations and phytase dose levels (up to a maximum of 3,000 FTU/kg) estimated that, across all phytases and studies, total tract P digestibility was increased by a maximum of 25 pp to a plateau at 64.9% (SE 1.82) (Rosenfelder-Kuon et al., 2020b).

Several previous studies have reported improvements in Ca digestibility by phytase in pigs (Kühn and Männer, 2012; Almeida et al., 2013; Adedokun et al., 2015), with the mode of action thought to be related to increased degradation of dietary phytate and reduced binding with Ca (both dietary and endogenous), thereby increasing Ca bioavailability for absorption. Consistent with those studies, in the present study, increasing phytase dose level (independent of limestone solubility) improved ATTD of Ca (g/kg basis) in an exponential manner. A previous study of PhyG in grower–finisher pigs fed corn-soybean meal-based diets reported no effect of the phytase on ATTD of Ca (Velayudhan et al., 2021). However, in that study the maximum phytase dose level tested was 1,000 FTU/kg, so dose-response effects may not have been so apparent. In addition, the ratio of Ca to (total) P in the diets was higher than in the present study (1.6 to 1.7:1 compared with 1.3:1) so that Ca would have been more likely to be in excess relative to P which may have reduced (in relative terms) the beneficial effect of the phytase.

The absence of an interaction suggested that limestone solubility did not affect the capacity of the phytase at increasing dose levels to improve P digestibility. In theory, the lower ATTD P in HSL diets could have been caused by a higher Ca content of the diet or imbalance with P, since it is well-established that Ca presence at greater than 2.15 × STTD P can reduce P digestibility in pigs (Selle et al., 2009; NRC, 2012). Indeed, analyzed Ca levels in the NC diets were slightly (0.16 to 0.17 pp) higher than formulated values, but this was consistent across HSL and MSL treatments. Further, analyzed P levels were very similar and close to formulated values and analyzed Ca:P ratios were equal across the two treatments (1.53 in both cases). Therefore, it seems unlikely that the lower ATTD P in diets containing HSL was linked to a difference in the Ca or P content of the diets and was more likely due to the solubility of the Ca present. The exponential curves for the relationship between ATTD of P (g/kg) against analyzed phytase activity (above NC) illustrated a lower ATTD P with a steeper slope for HSL, meaning a greater increase in response in HSL than MSL treatments, leading to a smaller difference between limestone solubilities at a high phytase dose of 2,000 FTU/kg. This suggests that increasing the phytase dose level reduced the negative impact of HSL on P digestibility. Previous studies have not generally considered the impact of limestone solubility on the phytase dose-response, even in broilers where more studies are available. In diets supplemented with a Buttiauxella sp. phytase (but without consideration of dose-response effects), Kim et al. (2018) reported that fine limestone reduced the AID of P at 28 d of age in broilers compared with a coarse limestone. Davin et al. (2020) reported a similar effect at 21 d of age in broilers fed diets containing fast- compared with slow-soluble limestone supplemented with either a Buttiauxella sp. or an E. coli phytase. The present results suggest that in young growing pigs fed diets containing HSL, it may be necessary to increase the dose level of PhyG phytase to meet animal P requirements.

Total tract digestibility of Ca (% and g/kg) was reduced in HSL compared with MSL diets containing phytase. As for effects on P digestibility, there was no interaction between phytase dose level and limestone solubility. In fact, even though the base level ATTD Ca (g/kg) in the NC diet without phytase was lower in the HSL treatments, the exponential curve fitting of the dose-response relationship showed a greater response with HSL than MSL, as indicated by the increased growth rate of the curve.

Limestone solubility effects on bone ash content have not frequently been assessed in pigs. In broilers, Davin et al. (2020) observed no effect on tibia ash at 21 d of age in diets containing Buttiauxella sp. phytase, whilst Majeed et al. (2020) reported no effect of limestone particle size on bone ash at either 14 or 35 d of age in diets supplemented with a Aspergillus oryzae phytase. By contrast, Manangi and Coon (2007) reported increased tibia ash content (mg of ash) in chickens fed with limestone of intermediate particle size (137–388 μm) compared to those with small (28 μm) or large (1306 μm) particle sizes. The findings of the present study are broadly consistent with those of Davin et al. (2020) and Majeed et al. (2020); increasing the dose level of PhyG between 0 and 2,000 FTU/kg improved bone ash content (%) in an exponential manner but bone ash was unaffected by limestone solubility. This suggests that, in the tested setting, the observed limestone solubility effects on mineral digestibility (ATTD of P and Ca), were not great enough to impair utilization of these minerals in bone. Indeed, at a dose level of 500 FTU/kg or above, bone ash in both MSL and HSL treatments was maintained equivalent to PC1 that contained 0.33% STTD of P. The estimated digestible P equivalence values (g/kg) of the phytase at a dose level of 1,000 FTU/kg in diets containing HSL are broadly consistent with the previous estimate of 1.83 g/kg digestible P from MCP made by Dersjant-Li et al. (2020), and higher than the range of values reported across multiple phytase studies in pigs by Dersjant-Li et al. (2015).

In conclusion, this study has generated the first data in pigs showing an effect of limestone solubility on the total tract digestibility of P and Ca. These effects were evident in diets without and with a novel consensus bacterial 6-phytase variant (PhyG). Across limestones, increasing phytase dose level from 0 to 2000 FTU/kg increased ATTD P exponentially. At 1,000 FTU/kg, the phytase was estimated to have released 1.89 g/kg and 2.32 g/kg of digestible P from MCP in medium- and high-solubility limestone diets, respectively, based on bone ash content. The results indicated that it may be necessary to increase the dose level of PhyG in diets containing HSL to achieve equivalent mineral digestibility and P-release improvements to those from diets containing less soluble limestone. More studies are needed to evaluate relationships between limestone solubility, mineral digestibility, and phytase efficacy in pigs.

Glossary

Abbreviations

AID

apparent ileal digestibility

ATTD

apparent total tract digestibility

DM

dry matter

FTU

phytase units

GMD

gross mean diameter

HSL

high-solubility limestone

IP₆

myo-inositol hexakisphosphate

ICP-OES

Inductively Coupled Plasma-Optical Emission Spectroscopy

MSL

medium solubility limestone

MCP

monocalcium phosphate

pp

percentage points

Pi

inorganic phosphate

PhyG

novel consensus bacterial 6-phytase variant

STTD

standardized total tract digestible

Conflict of Interest Statement

D.E.V., Y.D.-L., and L.M. are employees of Danisco Animal Nutrition & Health (IFF), a global supplier of microbial phytase. A.K. has no real or perceived conflicts of interest.