Comparison of relative calcium bioavailability based on bone ash and apparent ileal Ca digestibility in broiler chickens

Abstract

Two experiments were conducted to determine relative Ca bioavailability based on bone ash and apparent ileal digestibility (AID) of Ca in reagent-grade calcium carbonate (RCaCO₃) and RCaCO₃ + phytic acid in commercial broiler chickens. In Experiment 1, relative Ca bioavailability of RCaCO₃ and RCaCO₃ + phytic acid was determined in ad libitum-fed broiler chickens using the slope-ratio method. Dietary treatments were corn-soybean meal-based diets. Diet 1 was Ca-deficient (0.30 % Ca; 0.45 % non-phytate P), while diets 2 through 5 were the same as diet 1, except that 2 increasing levels of RCaCO₃ or RCaCO₃ + phytic acid were added to increase the dietary Ca level to 0.45 or 0.60 %. In Experiment 2, AID of Ca in RCaCO₃ with and without phytic acid was determined in ad libitum-fed broiler chickens using corn-based diets (0.83 % Ca and 0.45 % non-phytate P). There were 2 dietary treatments that were identical, except diet 1 contained RCaCO₃ while diet 2 contained RCaCO₃ + phytic acid. In Experiments 1 and 2, there were 5 and 8 replicate pens of 5 chickens per treatment, respectively. In Experiment 1, tibia ash (% and mg/tibia) increased linearly (P < 0.05) as dietary Ca content increased. Relative Ca bioavailability obtained from multiple linear regression of tibia ash content (mg/tibia) on supplemental Ca intake was reduced from 100 to 80 % (P < 0.05) when phytic acid was added. In Experiment 2, phytic acid reduced AID of Ca in RCaCO₃ from 38 to 30 % and relative AID of Ca was reduced to 80 %. In conclusion, phytic acid reduced both relative Ca bioavailability based on bone ash and AID of Ca by 20 %. Furthermore, the relative comparison between the 2 test ingredients in both bioassays was similar, indicating that both assays can potentially be used to assess Ca availability and that Ca digestibility values may be able to be predicted from relative bioavailability values as long as the AID of Ca in the standard reference Ca source is known.

Introduction

Calcium is the most abundant mineral in the body and has crucial roles in skeletal formation and maintenance, enzyme activation, and muscle contraction (Proszkowiec-Weglarz and Angel, 2013; Veum, 2010). Calcium deficiency can cause developmental issues with bone mineralization, whereas excess dietary Ca can reduce mineral availability and exacerbate infections (Paiva et al., 2013; Walk and Rao, 2020). It is, therefore, important that the Ca requirement of birds is met while minimizing excess dietary Ca levels. Current diet formulation is based on total Ca requirements, which does not account for differences in Ca availability among sources. This is important to consider, as Ca digestibility has been reported to vary among feedstuffs (Anwar et al., 2017, 2018; David et al., 2019; Dilworth et al., 1964; Walk et al., 2021). Further, Ca availability among different sources of the same ingredient can vary due to factors such as solubility and particle size (Walk et al., 2012, 2021). Therefore, diet formulation based on available Ca rather than total Ca may improve the precision by which the Ca requirement of poultry is met and allow nutritionists to ameliorate negative effects from excess dietary Ca.

One way to assess Ca availability is relative Ca bioavailability, where multiple levels of a standard reference ingredient and test ingredients are supplemented (Drysdale et al., 2024; Ward et al., 1984). The relative Ca bioavailability is then calculated using the slope-ratio method (Finney, 1964; Littell et al., 1995). Frequently used response variables for this type of bioassay include bone ash concentration, total bone ash content, bone density, and bone-breaking strength and standard reference ingredients include feedstuffs such as reagent-grade CaCO₃ (RCaCO₃) and limestone (Drysdale et al., 2024; Rao et al., 1993; Ward et al., 1984). The Ca bioavailability in standard reference ingredients is assumed to be 100 % when relative Ca bioavailability of test ingredients is calculated; however, it has been shown that Ca digestibility in limestone is variable, where Walk et al. (2021) reported Ca digestibility of limestone ranged from 22 to 77 %. Thus, although relative Ca bioavailability assays based on bone ash may work well for relative comparisons, a classical limitation of this type of bioassay is that values are not quantitative as there is no known standard ingredient that has a Ca digestibility value of 100 % (Walk et al., 2021).

In addition to relative Ca bioavailability, ileal Ca digestibility can also be used to examine the ability of birds to absorb the Ca in feed ingredients. There is increased interest in formulating poultry diets based on digestible Ca rather than total as discussed previously. This has been a challenge, however, as Ca digestibility can be highly variable among experiments due to ingredient variability, variations in feedstuff particle size, and differences in methodology such as dietary Ca to P ratios, the presence or absence of phytate, and the age of the birds used (Walk et al., 2021). Further, it can be difficult to obtain accurate analytical results for Ca in the diet and ileal digesta due to the lack of homogenous dispersion of Ca particles in samples and the small sample size used for analyses (Cromwell et al., 2003; David et al., 2019). Despite some of these limitations, a major advantage of ileal Ca digestibility assays over more classical relative bioavailability assays based on bone ash is that they provide a direct and quantitative assessment of Ca absorption by the bird (Drysdale et al., 2024; WPSA, 2013). These assays can also be conducted using short experimental lengths, as short as 24 h or less, compared with relative bioavailability studies that are approximately 14 d in length. Due to the concerns regarding adaptations by the birds to dietary Ca levels, it has been hypothesized that the ileal digestibility assays may be more accurate as potential adaptations by the birds to the dietary treatments can be circumvented by reducing the time in which diets are fed (David et al., 2019). Another advantage of ileal digestibility studies is that similar to P digestibility, Ca digestibility studies can be conducted using fewer dietary treatments than relative bioavailability studies based on bone ash.

The objective of this study was to evaluate the effect of phytic acid on Ca availability using 2 bioassays and to compare relative Ca bioavailability based on bone ash with apparent ileal digestibility (AID) of Ca using 2 samples that differed in Ca availability. It was hypothesized that phytic acid would reduce Ca availability and both bioassays would yield similar results.

Materials and methods

The protocol for this study was reviewed and approved by the Institutional Animal Care and Use Committee at the University of Arkansas (number 24011).

Diets and design

Experiment 1 was conducted to determine relative Ca bioavailability in feedstuffs using bone ash as the primary response criterion. Cobb 500 male broiler chickens were used in Experiment 1. Broiler chickens for Experiments 1 and 2 were obtained from a commercial hatchery and were housed in batteries with raised wire floors in an environmentally controlled room (cages were 24 in by 24 in). Chickens were provided a nutritionally complete crumbled corn-soybean meal starter diet until 7 d-of-age. Prior to the start of the experimental phase, birds were fasted overnight. On d 7, broilers were weighed and allotted to dietary treatments. A completely randomized design was used, and birds were allotted to achieve a similar mean BW among treatments. There were 5 dietary treatments (mash diets) that were provided to birds ad libitum from 7 to 21 d-of-age. On d 21, feed intake and BW were recorded per pen and birds were euthanized using CO₂ gas. The right leg was collected from each chicken to determine tibia ash content (mg/tibia) and concentration (%). The right tibia was autoclaved, cleaned of adhering tissues, oven-dried at 100°C for 24 h, and ashed at 600°C in a muffle furnace for 24 h. After ashed bones were weighed, bones were ground using a mortar and pestle and Ca in bone ash was measured to determine bone ash Ca concentration (%) and bone Ca content (mg/tibia). Multiple linear regression was conducted according to Finney (1964) and Littell et al. (1995) using the model below.

$$Y=a_0+b_1{x}_1+b_2{x}_2$$

Where: Υ = tibia ash concentration (%), tibia ash content (mg/tibia), or bone Ca content (mg/tibia); $a_0$ = y-intercept; $b_1$ = slope for RCaCO₃; $x_1$ = Ca intake (g) from RCaCO₃; $b_2$ = slope for RCaCO₃ + phytic acid; $x_2$ = Ca intake (g) from RCaCO₃ + phytic acid.

The 5 dietary treatments in Experiment 1 were corn soybean meal-based mash diets (Table 1). Diet 1 was a Ca-deficient diet which was formulated to contain 0.30 % Ca and was adequate in all other nutrients. Diets 2 and 3 were the same as diet 1, except that 0.37 % and 0.74 % RCaCO₃ was added in place of Solka Floc, respectively. Diets 4 and 5 were the same as diets 2 and 3, except that 0.15 % and 0.30 % phytic acid (myo-inositol 1,2,3,4,5,6-hexakis) was added, respectively. Diets 2 and 4 had a calculated Ca content of 0.45 % and diets 3 and 5 had a calculated Ca content of 0.60 %. These Ca levels were previously validated in the study by Drysdale et al. (2024) in commercial broiler chickens to confirm that they would yield linear responses in bone ash as the dietary Ca level increased. All diets contained a calculated non-phytate P content of 0.45 % and a calculated CP content of 22 %. Phytic acid was added to diets herein to reduce Ca availability in RCaCO₃ to provide 2 ingredients with differing Ca availability so that relative Ca bioavailability based on bone ash (Experiment 1) and relative AID of Ca (Experiment 2) could be compared. Further, a mixture of corn and RCaCO₃ ± phytic acid was made and the same mixture was used for diets in both Experiments 1 and 2 to ensure the same samples were being compared between bioassays. The amount of phytic acid mixed with corn was done to yield phytic acid levels similar to what is found in high phytate ingredients, such as wheat middlings. The RCaCO₃ used herein was obtained from Fisher Scientific (CAS: 471-34-1) and phytic acid solution (50 % w/w in H₂O) was obtained from Millipore Sigma (CAS: 83-86-3). The physical characteristics of RCaCO3 herein are as follows: density = 0.9 g/cm³; geometric mean diameter = 83.0 μm; geometric standard deviation = 1.7 μm).

Table 1.

Ingredient composition of diets in Experiment 1 (%, as-fed basis).

	Dietary treatments
Ingredient	1	2	3	4	5
Corn	59.00	59.00	59.00	58.85	58.58
Soybean meal	36.00	36.00	36.00	36.00	36.00
Soybean oil	1.40	1.40	1.40	1.40	1.40
Reagent-grade CaCO3	-	0.37	0.74	0.37	0.74
Phytic acid solution1	-	-	-	0.15	0.30
Dicalcium phosphate	0.75	0.75	0.75	0.75	0.75
Monosodium phosphate	0.72	0.72	0.72	0.72	0.72
Potassium chloride	0.02	0.02	0.02	0.02	0.02
Solka Floc2	1.04	0.67	0.30	0.67	0.39
L-Lys HCl	0.20	0.20	0.20	0.20	0.20
DL-Met	0.34	0.34	0.34	0.34	0.34
L-Thr	0.13	0.13	0.13	0.13	0.13
Vitamin mix3	0.05	0.05	0.05	0.05	0.05
Mineral mix4	0.10	0.10	0.10	0.10	0.10
NaCl	0.15	0.15	0.15	0.15	0.15
Choline chloride	0.10	0.10	0.10	0.10	0.10
Nutrients values
CP, calculated	21.7	21.7	21.7	21.7	21.7
Ca, calculated	0.30	0.45	0.60	0.45	0.60
Ca, analyzed	0.38	0.55	0.66	0.53	0.63
Non-phytate P, calculated	0.45	0.45	0.45	0.45	0.45
Total P, analyzed	0.74	0.78	0.75	0.77	0.80
Phytic acid, analyzed	1.53	1.55	1.46	1.62	1.66
Na, calculated	0.24	0.24	0.24	0.24	0.24

¹Phytic acid (myo-inositol 1,2,3,4,5,6-hexakis) solution was a 50/50 mixture with water.

²Powdered cellulose from International Corporation, Urbana, OH 43078.

³Provided per kilogram of diet: vitamin A, 6,173 IU; vitamin D3, 4,409 ICU; vitamin E, 44 IU; vitamin B12, 0.01 mg; menadione, 1.20 mg; riboflavin, 5.29 mg; d-pantothenic acid, 7.94 mg; thiamine, 1.23 mg; niacin, 30.86 mg; pyridoxine, 2.20 mg; folic acid, 0.71 mg; biotin, 0.07 mg.

⁴Provided as milligrams per kilogram of diet: manganese, 100 mg; zinc, 100 mg; iron, 15 mg; selenium, 0.25 mg; copper, 15 mg; iodine, 1.2 mg.

Experiment 2 was conducted to determine AID of Ca in feedstuffs. Commercial Cobb 500 male broiler chickens were housed in batteries with raised wire floors in an environmentally controlled room (cages were 24 in by 24 in). Chickens were provided with a nutritionally complete corn-soybean meal crumbled starter diet for 18 d and had ad libitum access to feed and water. On d 18, birds were fasted overnight. On d 19, chickens were weighed and allotted to 1 of 2 dietary treatments, equalizing BW across treatments in a completely randomized design. Experimental diets were provided ad libitum from d 19 to 21. Both dietary treatments were corn-based mash diets (93 % corn) that contained 2.02 % RCaCO₃ with or without 0.81 % of a phytic acid solution (Table 2). Each diet had a calculated non-phytate P content of 0.45 % and calculated CP level of 6.4 %. Diets 1 and 2 were formulated to contain 0.83 % total Ca. Titanium dioxide was added at 0.5 % of the diet as an indigestible marker. Chickens were euthanized on the final day of the experiment (21 d-of-age) using CO₂ gas and ileal digesta were collected from the distal half of the small intestine (Meckel’s diverticulum to the ileocecal junction) via flushing. Diets and freeze-dried ileal digesta were analyzed for Ca and Ti. The AID of Ca was calculated as described below.

AID of Ca (%) = [(Ca in diet (%) – Ca in digesta (%) x (Ti in diet (%) / Ti in digesta (%))) / Ca in diet] × 100

Table 2.

Ingredient composition of diets in Experiment 2 (%, as-fed basis).

	Dietary treatments
Ingredient	1	2
Corn	93.38	92.57
Soybean oil	2.00	2.00
Reagent-grade CaCO3	2.02	2.02
Phytic acid solution1	-	0.81
Monosodium phosphate	1.10	1.10
Potassium phosphate	0.40	0.40
KCl	0.35	0.35
Vitamin mix2	0.05	0.05
Mineral mix3	0.10	0.10
Choline chloride	0.10	0.10
Titanium dioxide	0.50	0.50
Nutrients
CP, calculated	6.40	6.40
Ca, calculated	0.83	0.83
Ca, analyzed	0.89	0.84
Non-phytate P, calculated	0.45	0.45
Total P, analyzed	0.78	0.94
Phytic acid, analyzed	1.50	1.88
Na, calculated	0.23	0.23

¹Phytic acid (myo-inositol 1,2,3,4,5,6-hexakis) solution was a 50/50 mixture with water.

²Provided per kilogram of diet: vitamin A, 6,173 IU; vitamin D3, 4,409 ICU; vitamin E, 44 IU; vitamin B12, 0.01 mg; menadione, 1.20 mg; riboflavin, 5.29 mg; d-pantothenic acid, 7.94 mg; thiamine, 1.23 mg; niacin, 30.86 mg; pyridoxine, 2.20 mg; folic acid, 0.71 mg; biotin, 0.07 mg.

³Provided as milligrams per kilogram of diet: manganese, 100 mg; zinc, 100 mg; iron, 15 mg; selenium, 0.25 mg; copper, 15 mg; iodine, 1.2 mg.

Chemical analyses

Titanium concentrations in experimental diets and ileal digesta were measured using UV spectroscopy (Myers et al., 2004). The Ca and P analyses were performed using inductively coupled plasma-mass spectrometry (method 985.01 A, B, and C; AOAC International, 2007). Bone ash, Ca, and P analyses were conducted at the University of Arkansas Central Analytic Lab, Fayetteville, AR. Titanium was measured at University of Missouri-Columbia Experimental Station Chemical Laboratory, Columbia, MO.

Statistical analysis

SAS software (SAS Institute INC, 2010) was used for all statistical analyses. Data from Experiments 1 and 2 were analyzed using a 1-way ANOVA for a completely randomized design. Multiple linear regression analysis was then conducted using data from Experiment 1 to determine relative Ca bioavailability. For this analysis, bone ash concentration (%), bone ash content (mg/tibia), and bone Ca content (mg/tibia) were regressed on supplemental Ca intake from RCaCO₃ and RCaCO₃ + phytic acid (Finney, 1964). The relative bioavailability of Ca in RCaCO₃ with and without phytic acid was calculated using the slope ratio method (Dilworth et al., 1964). The Ca bioavailability value for RCaCO₃ was set at 100 % and the regression coefficient (or slope) for RCaCO₃ + phytic acid was divided by the regression coefficient (or slope) for RCaCO_3. Experiment 1 had 5 replicate pens with 5 chickens per pen and Experiment 2 had 8 replicate pens with 5 chickens per pen, which served as the experimental unit. For multiple linear regression analyses in Experiment 1, each pen was used as an observation rather than the treatment mean due to the limited number of dietary treatments per test ingredient to ensure that the responses did not deviate from linearity. Differences among means were evaluated using Fisher’s least significant difference test (Carmer and Walker, 1985) and slope contrasts were used to assess differences among relative Ca bioavailability values in Experiment 1. The significance value for all analyses was P < 0.05.

Results and discussion

Experiment 1

Growth performance results from Experiment 1 are presented in Table 3. The BW gain and gain: feed were lower (P < 0.05) for treatment 1 compared with 2 through 5. Feed intake was generally lower (P < 0.05) for treatment 1 compared with other treatments. It is interesting to note that Drysdale et al. (2024) reported increases in feed intake in response to the addition of dietary Ca using similar diets to those herein for both cross-bred and Ross 308 broiler chickens. Similar comparisons have also been reported for P bioavailability studies, where cross-bred and Ross 308 broiler chickens consumed more feed as P was added to P-deficient diets, while few differences in feed intake were observed in Cobb 500 broilers (Hanna et al., 2018; Parsons et al., 2023a; Parsons and Rochell, 2024). This suggests there are breed-specific responses in feed intake to mineral-deficient diets. Overall, there was no difference (P > 0.05) among the test ingredients for BW gain, feed intake, and gain: feed. Dry bone weight (mg/tibia) and bone ash concentration (%) increased (P < 0.05) as supplemental RCaCO₃ increased (Table 4). For example, bone ash concentration increased from 42 % for treatment 1 to 46 and 45 % for treatments 3 and 5, respectively. Bone ash content (mg/tibia) followed similar trends, where there was an increase (P < 0.05) from 734 mg/tibia for treatment 1 to 1018 and 967 mg/tibia with the highest inclusion of RCaCO₃ and RCaCO₃ + phytic acid, respectively. Similar to bone ash content (mg/tibia), bone Ca content increased as the dietary Ca level increased (P < 0.05). In contrast to other bone response variables, however, there was no linear effect of dietary Ca level on bone Ca concentration (P > 0.05); therefore, multiple linear regression analysis was not conducted using this response variable as it deviated from linearity.

Table 3.

Growth performance of chickens in Experiment 1¹.

Dietary treatment	BW gain (g/chicken)	Feed intake (g/chicken)	Gain:feed (g/kg)
1. Ca deficient diet - 0.30 % Ca	693b	1012b	685b
2. As 1 + 0.37 % Ca from RCaCO3	785a	1068a	734a
3. As 1 + 0.74 % Ca from RCaCO3	787a	1087a	723a
4. As 1 + 0.37 % Ca from RCaCO3 + 0.075 % phytic acid	759a	1054ab	720a
5. As 1 + 0.74 % Ca from RCaCO3 + 0.150 % phytic acid	776a	1084a	716a
SEM	16.2	14.4	8.1
P-value	<0.001	0.009	0.005

^a-bMeans within a column with no common superscript differ (P < 0.05). Values are means of 5 pens of 5 chickens from 7 to 21 d-of-age.

¹Abbreviations: RCaCO₃ = reagent-grade CaCO₃; phytic acid = myo-inositol 1,2,3,4,5,6-hexakis.

Table 4.

Tibia ash concentration (%), tibia ash content (mg/tibia), and tibia Ca content (mg/tibia) of chickens in Experiment 1¹.

Dietary treatment	Dry bone weight (mg/tibia)	Bone ash2 (%)	Bone ash3 (mg/tibia)	Bone Ca (%)	Bone Ca4 (mg/tibia)
1. Ca deficient diet - 0.30 % Ca	1746c	42.1c	734c	32.1b	236d
2. As 1 + 0.37 % Ca from RCaCO3	1963b	44.3b	870b	33.6a	292c
3. As 1 + 0.74 % Ca from RCaCO3	2199a	46.3a	1018a	33.8a	359a
4. As 1 + 0.37 % Ca from RCaCO3 + 0.075 % phytic acid	1943b	44.1b	856b	33.1ab	287c
5. As 1 + 0.74 % Ca from RCaCO3 + 0.150 % phytic acid	2139a	45.2a	967a	32.2b	311b
SEM	35.9	0.27	14.8	0.36	6.0
P-value	<0.001	<0.001	<0.001	0.008	<0.001

^a-dMeans within a column with no common superscript differ (P < 0.05). Values are means of 5 pens of 5 chickens from 7 to 21 d-of-age.

¹Abbreviations: RCaCO₃ = reagent-grade CaCO₃; phytic acid = myo-inositol 1,2,3,4,5,6-hexakis.

²Multiple regression of tibia ash (Y; %) on supplemental Ca intake (g) from RCaCO₃ (X₁), and RCaCO₃ + phytic acid (X₂) yielded the equation: Y = 42.23 + 1.32 ± 0.12 X₁ + 0.96 ± 0.10 X₂ (R² = 0.89 and RMSE = 0.55). The (±) values are standard errors of the regression coefficients.

³Multiple regression of tibia ash (Y; mg) on supplemental Ca intake (g) from RCaCO₃ (X₁), and RCaCO₃ + phytic acid (X₂) yielded the equation: Y = 731.8 + 92.2 ± 5.6 X₁ + 73.9 ± 5.3 X₂ (R² = 0.94 and RMSE = 35.3). The (±) values are standard errors of the regression coefficients.

⁴Multiple regression of tibia ash Ca (Y; mg) on supplemental Ca intake (g) from RCaCO₃ (X₁), and RCaCO₃ + phytic acid (X₂) yielded the equation: Y = 240.7 + 33.1 ± 2.9 X₁ + 23.1 ± 2.9 X₂ (R² = 0.86 and RMSE = 15.8). The (±) values are standard errors of the regression coefficients.

Multiple linear regression of tibia ash concentration (%), tibia ash content (mg/tibia), and tibia Ca content (mg/tibia) versus supplemental Ca intake from RCaCO₃ are presented in Fig. 1. There was a linear increase (P < 0.05) in both tibia ash concentration (%) and tibia ash content (mg/tibia) as supplemental Ca intake for both ingredients increased. Further, a high R² value of 0.89 was observed for regressions based on tibia ash concentration (%) and 0.94 for tibia ash content (mg/tibia). Multiple linear regression was also conducted using tibia Ca content (mg/tibia) as shown in Fig. 1. Similar to regressions conducted using tibia ash concentration and content, there was a linear increase in tibia Ca content (mg/tibia) as supplemental Ca intake increased (P < 0.05; R² = 0.86). The lower R² value of 0.86 obtained when regressions were conducted using tibia Ca content compared with tibia ash content (R² = 0.94) is likely due to analytical errors associated with the measurement of Ca in the bone samples, such as non-uniform or homogenous distribution of Ca particles in the samples. The slope-ratio values obtained from Fig. 1 are shown in Table 5. Relative Ca bioavailability was reduced to 73 % for regressions based on tibia ash concentration (%), 80 % for tibia ash content (mg/tibia), and 70 % for tibia Ca content (%) when phytic acid was added. Based on the regression equations, standard errors of regression coefficients, and slope contrasts (Table 4 footnotes 2 through 4), all 3 response variables indicated that phytic acid reduced (P < 0.05) Ca bioavailability in RCaCO_3.

Fig. 1.

(A) Tibia ash concentration (%) vs. supplemental Ca intake from reagent-grade CaCO₃ (RCaCO₃; X₁) or RCaCO₃ + myo-inositol 1,2,3,4,5,6-hexakis (phytic acid; X₂). Linear regression yielded the following equation: $Y=42.23+1.32X_1+0.96X_2$ (R² = 0.89; RMSE = 0.55). (B) Tibia ash content (mg/tibia) vs. supplemental Ca intake from RCaCO₃ (X₁) or RCaCO₃ + phytic acid (X₂). Linear regression yielded the following equation: $Y=731.8+92.2X_1+73.9X_2$ (R² = 0.94; RMSE = 35.3). (C) Tibia Ca content (mg/tibia) vs. supplemental Ca intake from RCaCO₃ (X₁) or RCaCO₃ + phytic acid (X₂). Linear regression yielded the following equation: $Y=240.7+33.1X_1+23.1X_2$ (R² = 0.86; RMSE = 15.8).

Table 5.

Relative Ca bioavailability from 7 to 21 days-of-age in Experiment 1¹.

Sample	Relative Ca bioavailability values2 (%)
	Tibia ash concentration (%)	Tibia ash content (mg)	Tibia Ca content (mg)
RCaCO3	100a	100a	100a
RCaCO3 + phytic acid	73b	80b	70b
P-value	0.002	0.002	0.002

^a-bValues within a column with no common superscript differ (P < 0.05). Bioavailability values were compared using slope contrasts.

¹Abbreviations: RCaCO₃ = reagent-grade CaCO₃; phytic acid = myo-inositol 1,2,3,4,5,6-hexakis.

²Calculated by the slope-ratio method using the regression equation in Table 4 footnotes 2 through 4. Bioavailability values are relative to the Ca in RCaCO₃ which was set at 100 %.

Phytic acid is naturally found in plant-based feedstuffs and it is an important storage form of P for plants; however, it possesses anti-nutritional properties which can reduce nutrient availability beyond what is observed for P (Angel et al., 2002). It is well known that phytate P is primarily unavailable to chickens in feedstuffs with little endogenous feed phytase if an exogenous phytase is not used (Leske and Coon, 1999; Nelson, 1967; Parsons et al., 2023b). In addition to poor phytate P availability, when consumed by birds, negatively charged phytic acid or phytate can bind positively charged minerals in the diet, such as Ca, thereby reducing both P availability and the availability of the positively charged minerals (Angel et al., 2002; Kim et al., 2018; Liu et al., 2013; Parsons et al., 2024). Sommerfeld et al. (2019) investigated this in a study using gnotobiotic broilers chickens. These authors reported that when broiler chickens were fed low Ca corn-soybean meal-based diets (0.62 % Ca), 42 % of the phytic acid was degraded; however, phytic acid degradation was reduced to 17 % when the Ca content was increased to 1.04 %. Similarly, Venter et al. (2024) reported reductions in both Ca and P digestibility from 49 to 33 % and 31 to 16 %, respectively, as the phytate content in diets were increased from 0.57 to 1.22 %. These studies demonstrate that as dietary Ca is increased, the ability of endogenous phytases produced by the birds to degrade phytic acid is reduced due to Ca-phytate interactions, subsequently reducing the availability of both Ca and P. This is in good agreement with the present study, where the addition of purified phytic acid reduced Ca bioavailability in Experiment 1 (Table 5).

Experiment 2

Growth performance and AID of Ca from Experiment 2 are presented in Table 6. There was no effect of phytic acid on BW gain or gain:feed (P > 0.05). It was observed, however, that feed intake of birds fed RCaCO₃ + phytic acid was greater (P < 0.05) compared with RCaCO₃ without phytic acid, being 255 and 237 g/chicken, respectively. The lack of effect of phytic acid on BW gain and gain:feed was expected due to the short duration of the experimental phase (48 h). The cause for greater feed intake when phytic acid was added remains to be elucidated, however, it is possible that the palatability of the feed may have been altered by the addition of phytic acid or that the birds consumed more feed as a response to the reduction in Ca availability caused by the addition of phytic acid. The AID of Ca was 38 % for RCaCO₃ and was reduced to 30 % for RCaCO₃ + phytic acid (P < 0.05; Table 6). The relative AID of Ca was then calculated using a similar method as relative bioavailability where the AID of Ca of RCaCO₃ + phytic acid was divided by the AID of Ca in RCaCO₃ without phytic acid and then multiplied by 100. The AID of Ca in the RCaCO₃ + phytic acid relative to RCaCO₃ was 80 %. Results from Experiment 2 were in good agreement with Experiment 1, where relative Ca bioavailability based on bone ash concentration (%), bone ash content (mg/tibia), and bone Ca content (mg/tibia) were all reduced by the addition of phytic acid (Table 5). Further, both bioassays yielded the same relative Ca availability estimate of 80 % for relative Ca bioavailability based on bone ash content (Experiment 1; Table 5) and relative AID of Ca (Experiment 2; Table 6).

Table 6.

Growth performance and apparent ileal calcium digestibility for chickens in Experiment 2.

Test ingredient	BW gain (g/chicken)	Feed intake (g/chicken)	Gain:feed (g/kg)	AID of Ca (%)	Relative Ca digestibility2 (%)
1. 2.02 % Ca from RCaCO3	124	237b	525	38a	100
2. 202 % Ca from RCaCO3 + 0.41 % phytic acid	126	255a	493	30b	80
SEM	5.5	5.4	13.5	1.5	-
P-value	0.849	0.032	0.113	0.002	-

^a-bMeans within a column with no common superscript differ (P < 0.05). Values are means of 8 pens of 5 chickens from 19 to 21 days-of-age.

¹Abbreviations: AID = apparent ileal digestibility; RCaCO₃ = reagent-grade CaCO₃; phytic acid = myo-inositol 1,2,3,4,5,6-hexakis.

²Bioavailability values are relative to the Ca in RCaCO₃ which was set at 100 %.

The AID of Ca in RCaCO₃ herein was in good agreement with Kim et al. (2018) who reported that pulverized limestone had an AID of Ca of 38 % in a corn-soybean meal-based diet. In a review conducted by Walk et al. (2021), 55 samples of limestone were evaluated and the mean AID of Ca was 53 %. The authors reported in that review, however, that there was a large variation in Ca digestibility values among the 55 limestones, ranging from 20 to 77 %. The low AID of Ca for RCaCO₃ of 38 % in the present study may be due to the use of corn-based diets without exogenous phytase. As discussed above and as observed in Experiments 1 and 2 herein, phytate can bind to Ca and prevent its absorption by the bird (Kim et al., 2018; Venter et al., 2024). Further, the rapidly soluble RCaCO₃ also likely exacerbated the Ca-phytate interactions contributing to the lower AID of Ca compared with the limestone samples evaluated by Walk et al. (2021). Had a phytate-free diet been used or exogenous phytase been added, it is likely that the AID of Ca in RCaCO₃ would have been greater than the value of 38 % in the present study.

Another potential contributing factor to the low AID of Ca in RCaCO₃ is dietary adaptations by the birds. It has been reported that Ca digestibility values are influenced by the length of the experimental phase. For example, David et al. (2019) reported large reductions in the AID of limestone from 65 to 36 % as the dietary feeding length increased from 24 to 168 h, with the greatest reduction occurring between 24 and 72 h. It has been hypothesized that the reduction in Ca absorption over time may be due to down-regulation of Ca transporters. It should be noted, however, that both the dietary Ca level during the pre-experimental phase and during the experimental phase (0.64 vs. 0.84 %) have been evaluated and neither were found to affect the AID of Ca in RCaCO₃ (B. Parsons, unpublished data). Another potential explanation for the reduced digestibility of Ca over time reported by David et al. (2019) is that Ca was retained in the gizzard and had not reached a steady state of equilibrium flowing through the GI tract, making it appear as though birds absorbed more dietary Ca than they actually did. David et al. (2021) reported that broiler chickens fed diets for 3 d had substantially greater Ca concentrations in the gizzard when fed a coarse limestone (1301 µm) compared with a fine limestone (462 µm), being 4.6 % Ca and 0.38 % Ca, respectively. The proposition that birds are highly responsive in adapting to dietary Ca levels would be a limitation of relative bioavailability assays based on bone ash as the bioassays must be conducted for approximately 14 days to allow sufficient time for Ca to be deposited in the bone. It is important to note, however, that linear responses in bone ash to increasing dietary Ca levels were observed herein and previously by other authors (Drysdale et al., 2024; Ward et al., 1984). Further, the negative effect of phytic acid on Ca availability was observed in both Experiments 1 and 2 which had 14 d and 48 h feeding lengths, respectively. This is interesting to consider as adaptations to dietary Ca by the birds would have likely yielded non-linear responses from multiple linear regression and prevented the detection of differences in Ca availability between the 2 test ingredients in Experiment 1; thus, data herein suggest there may have been little to no adaptation to dietary Ca levels by the birds in this study.

There has been little work comparing relative Ca bioavailability based on bone ash with AID of Ca. In a previous study, Drysdale et al. (2024) evaluated 5 limestones of various origins using both a relative Ca bioavailability assay based on bone ash and an ileal Ca digestibility assay. In that study, relative Ca bioavailability based on bone ash was high (90 to 96 %) compared with AID of Ca (20 to 34 %). This is in good agreement with the present study where the bioavailability of RCaCO₃ + phytic acid relative to RCaCO₃ without phytic acid was higher (80 %) compared with absolute AID of Ca (30 %). To the best of the authors knowledge, there is no known Ca source that is 100 % digestible, thus highlighting the importance of knowing the AID of Ca in standard reference ingredients used in relative bioavailability studies. Although few comparisons between bioassays could be made in the study by Drysdale et al. (2024) due to a lack of differences in Ca availability among limestones in the study, the 2 test ingredients evaluate herein clearly differed in Ca availability and thus the relative bioavailability based on bone ash could be compared with relative AID of Ca. In fact, the addition of phytic acid in the present study was conducted for the exact purpose of generating 2 ingredients that would differ in Ca availability to provide a clear comparison between the two bioassays. Further, as mentioned previously, it can be difficult to obtain accurate Ca analyses in samples due to non-homogeneous dispersion of Ca or limestone particles (Cromwell et al., 2003). The RCaCO₃ used as the test source of Ca in Experiments 1 and 2, however, was a fine powder which the authors had previously found is more accurately detected in samples compared with limestone (unpublished data). Data herein demonstrated good agreement between the Ca bioavailability assay and ileal digestibility assay, where relative Ca bioavailability based on bone ash content (mg/tibia) in Experiment 1 was 80 % (Table 5), which was the same as the relative AID of Ca in Experiment 2 (Table 6). This demonstrates that relative Ca bioavailability based on bone ash may provide similar estimates to relative AID of Ca and thus further research is warranted to better define the relationship between Ca availability estimates obtained from these 2 bioassays.

The similarity between Ca availability estimates obtained from the relative bioavailability assay based on bone ash and the relative ileal digestibility assay herein has useful implications. This suggests that as long as the AID of Ca in the standard reference ingredient is known, researchers may be able to convert relative bioavailability values based on bone ash content (mg/tibia) to predicted or estimated Ca digestibility values. The digestibility assay is useful because it provides direct measurements of Ca availability. This bioassay is also shorter and can be conducted using a single treatment per test ingredient, which reduces the number of animals and treatments needed compared with the relative Ca bioavailability assays based on bone ash. Due to challenges associated with Ca analyses and homogenous dispersion of Ca particles in samples, however, Ca bioavailability assays based on bone ash can be helpful to confirm results from digestibility assays while also eliminating analytical errors. Further, relative bioavailability based on bone ash will account for absorption, transportation, and usage within the body, which offers additional insights beyond absorption or digestibility (Soares, 1995). Overall, both of these bioassays provide valuable information on feedstuffs and data herein demonstrates good agreement between the bioassays, thus providing researchers with another means by which to assess Ca availability in feedstuffs.

In summary, a slope-ratio relative Ca bioavailability assay based on bone ash and an ileal Ca digestibility assay were conducted using commercial broiler chickens. The availability of Ca in RCaCO₃ was determined with and without the addition of phytic acid to provide 2 test ingredients that differed in Ca availability. The addition of phytic acid reduced relative Ca bioavailability and relative AID of Ca. There was good agreement between bioassays, particularly for relative Ca bioavailability based on tibia ash content (mg/tibia) and AID of Ca, where the addition of phytic acid reduced relative Ca availability to 80 %. These results show that both the more classical relative bioavailability assay based on bone ash and the ileal Ca digestibility assay can be used by researchers when assessing Ca availability in feedstuffs and relative bioavailability values can provide an estimate of Ca digestibility if the AID of Ca in the standard reference ingredient is known.

Disclosures

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Benjamin W Parsons reports was provided by UNIVERSITY OF ARKANSAS. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.