Introduction
Macrominerals are essential components of ruminant metabolism, with calcium (Ca) required principally as a structural element, constituting the bones in a stoichiometric ratio with phosphorus (P) (Suttle, Reference Suttle2010). In addition, Ca participates in several intra- and intercellular processes such as nerve transmissions, muscle contraction and hormone regulation (Buttery et al., Reference Buttery, Brameld, Dawson and Cronjé2000). Calcium deficiency can lead to rickets in young animals, osteomalacia in adults and milk fever in lactating ewes (NRC 2007). Thus, the accurate estimation of Ca requirements is necessary to assure the proper functioning of maintenance and growth processes in animals.
The NRC (2007) reports Ca requirements for the maintenance of goats using data obtained partially from sheep, which could be inappropriate since serum and bone Ca metabolism differs between sheep and goats (Liesegang and Risteli, Reference Liesegang and Risteli2005). Previous studies demonstrated that under Ca dietary restriction, sheep decrease plasma Ca and increase phosphate concentrations, while goats compensate for low dietary Ca supply by stimulating vitamin D-dependent intestinal Ca absorption (Wilkens et al., Reference Wilkens2012, Reference Wilkens, Breves and Schröder2014). Several studies estimating the Ca requirements for maintenance have been conducted in goats (Fernandes et al., Reference Fernandes2012; Teixeira et al., Reference Teixeira2015; Santos et al., Reference Santos2016); however, the dynamics of Ca (i.e. true Ca intake, absorption, urine excretion, faeces excretion, retention and endogenous losses) have not been assessed so far. This is necessary to improve the knowledge of Ca metabolism and hence, Ca requirements for maintenance.
The dynamics of Ca can be assessed using the isotopic dilution method, which is a widely accepted technique for mineral dynamic studies in animals (Vitti and Kebreab, Reference Vitti and Kebreab2010). This method estimates the true Ca inputs and outputs in comparison to conventional assays (e.g. feeding trials and comparative slaughter studies), which are only able to obtain the apparent Ca inputs and outputs (Roque et al., Reference Roque2007; Fernandes et al., Reference Fernandes2012). Thus, the objectives of the current study were to investigate the dynamics of body Ca of goats fed diets with different Ca concentrations and to estimate the net Ca requirements for maintenance of Saanen goats, using 45Ca as a radiotracer.
Materials and methods
Humane animal care and handling procedures were conducted in accordance with the Animal Care Committee (Comissão de Ética e Bem Estar Animal) of the Sao Paulo State University and with instructions from the Ministry of Agriculture in Brazil (instruction number 56/2008).
Animal handling, experimental design and sampling
Eighteen castrated male Saanen goats (25 ± 2.3 kg body weight (BW), 180 days old) were housed indoors in metabolic cages (0.6 × 1.2 m2) designed for isotope studies, which allowed total collection of faeces and urine and the control of feed intake. The goats received a basal diet composed of ground ear maize, finely ground maize grain and vitamin–mineral premix (Table 1). The diets were formulated to meet the daily requirements for maintenance of 438 kJ/kg0.75 BW of metabolizable energy, 2.2 g of protein/kg0.75 BW and 479 mg/day of dietary P, according to AFRC (1998). All goats had free access to fresh water. The treatments consisted of adding different amounts of calcitic limestone to the basal diet (0.0, 3.0 and 7.0 g/kg dry matter (DM)), resulting in 0.6, 1.7 and 3.0 g Ca/kg DM, respectively (Table 1). These Ca levels were selected to be below Ca maintenance requirement, at Ca maintenance requirement and above Ca maintenance requirement, respectively. A constant P dietary supply (i.e. meeting the net P requirement of goats) was assured across diets to test changes of Ca dietary levels but not of P dietary levels on dynamics of body Ca.
Composition and characterization of experimental diets
DM, dry matter; CP, crude protein; ADF, acid detergent fibre; NDF, neutral detergent fibre; EE, ether extract; P, phosphorus.
a Diets: 0.6 = 0.6 g Calcium (Ca)/kg DM; 1.7 = 1.7 g Ca/kg DM; and 3.0 = 3.0 g Ca/kg DM.
b Ground ear maize included the cobs, grains and husks.
c Ground maize included grains only.
d Calcitic limestone (Ca = 367.5 g/kg DM; P = 95.2 mg/kg DM).
e Vitamin–mineral premix = Salt, mineral supplement and vitamin supplement. Micro mineral composition of Premix: Iron = 8.73 mg/kg DM; Copper = 7.64 mg/kg DM; Manganese = 44.36 mg/kg DM; Zinc = 58.73 mg/kg DM; Cobalt = 0.1 mg/kg DM; Iodine = 0.2 mg/kg DM; Selenium = 0.03 mg/kg DM.
The experiment lasted 45 days, which included a 36-day adaptation period and 9 days for sample collection (i.e. control of feed intake, collection of faeces, urine and blood). To avoid feed orts, the animals received 65 g/day of feed per unit of metabolic BW. Feed was offered twice daily at 08:00 and 16:00 h. After the morning meal of day 38, each animal was injected, via the right jugular vein, with a single dose of 7.4 MBq 45Ca in a 0.5 ml sterile isotonic saline solution (9 g/l sodium chloride (NaCl)). Blood samples from all animals were collected from the right jugular vein at 5 and 60 min after radiotracer injection.
From days 39 to 45, samples of blood, faeces and urine were collected daily assuming that during this period the Ca-specific activity in plasma and faeces was in a steady-state condition (Lofgreen and Kleiber, Reference Lofgreen and Kleiber1954). The Ca concentration in plasma was analysed immediately after collection. Total faeces and urine excretion were quantified daily and representative samples (0.10) of each were stored at −20 °C for later analysis.
Chemical analysis
Feed samples were analysed chemically to determine DM (AOAC 1990; method: 930.15), crude protein (AOAC 1990; method: 954.01) and ether extract (AOAC 1990; method: 920.39). Neutral detergent fibre values were determined with amylase and without sulphite (Van Soest, Reference Van Soest1994) and acid detergent fibre values were determined according to Goering et al. (Reference Goering1972). Phosphorus concentrations were determined using the colorimetric method (AOAC 1990; method: 965.17). Calcium concentrations in feed samples, faeces and urine were determined with atomic absorption spectrometry (AOAC 1990; method: 935.13).
Blood samples were analysed to determine the concentration of inorganic Ca in plasma. Briefly, after blood collection, samples (10 ml) were centrifuged (1800 g for 10 min) to achieve plasma separation. Nine millilitres of trichloroacetic acid (100 g/l) was added to 1 ml of plasma for protein precipitation. After centrifugation (1000 g for 10 min), inorganic Ca in plasma was determined by atomic absorption spectrometry (AOAC 1990, method: 935.13).
Radioactivity in plasma, faeces and urine was determined using the liquid scintillation method (Passo and Cook, Reference Passo and Cook1994) within 15 days of radiotracer injection. Plasma and urine samples (1 ml each) were added to 10 ml of scintillation solution in borosilicate vials. Ashes of faecal samples (1 g) were previously dissolved in 5 ml of 18 N sulphuric acid (H2SO4) and then 1 ml was placed into borosilicate vials with 10 ml of scintillation solution. The composition of the scintillation solution was 4 g of 2,5-diphenyloxazole and 0.3 g of 1,4-bis-(5-phenyloxazole-2-il) benzene dissolved in a mixture of 500 ml Triton X-100 and 1 litre of toluene. The corrections for decay (45Ca half-life = 165 days) were carried out in relation to the standard prepared at injection.
Calculations
The proportions of injected activity of Ca in faeces (CaIAfc) and plasma (CaIApl) were determined according to Lofgreen and Kleiber (Reference Lofgreen and Kleiber1954) as follows:
$${\rm CaIAfc} = \; \displaystyle{{{\rm cpmfc}} \over {{\rm cpmsp}}}\; $$
$${\rm CaIApl} = {\rm \;} \displaystyle{{{\rm cpmpl}} \over {{\rm cpmsp}}}\; $$where cpmfc, cpmpl and cpmsp are the counts/min in 1 g of faeces, counts/min in 1 ml of plasma and counts/min in 0.5 ml of the standard solution of 45Ca containing 55 000 dpm/0.5 ml, respectively. The standard solution was prepared from the standard 45CaCl2 of 62.7 × 106 dpm/μl and ultra-pure water.
The specific activity of Ca in faeces (SAfc) and plasma (SApl) was determined by the following equations:
$${\rm SAfc} = \; \displaystyle{{{\rm CaIAfc}} \over {{\rm faeces}\,{\rm (mg}\,{\rm Ca}/{\rm g)}}}$$
$$\hskip11pt{\rm SApl} = {\rm \;} \displaystyle{{{\rm CaIApl}} \over {{\rm plasma}\,{\rm (mg}\,{\rm Ca}/{\rm ml)}}}$$where faeces and plasma are the Ca in faeces (mg Ca in 1 g faeces) and Ca in plasma (mg Ca in 1 ml plasma), respectively.
Using SAfc and SApl values, the proportion of faecal endogenous Ca (pCafce) was calculated using the equation:
$${\rm pC}{\rm a}_{{\rm fce}} = \; \displaystyle{{{\rm SAfc}} \over {{\rm SApl}}}\; $$The faecal Ca output (Cafc) and Cafce were used to calculate the total amount of faecal endogenous Ca (Cafce), according to the equation:
$${\rm C}{\rm a}_{{\rm fce}}({\rm g}/{\rm day}) = \; {\rm C}{\rm a}_{{\rm fc}}({\rm g}/{\rm day}){\rm \;} \times {\rm pC}{\rm a}_{{\rm fce}}$$The true absorbed Ca (Caabs) was determined using the following equation:
$$\eqalign{{\rm C}{\rm a}_{{\rm abs}}({\rm g}/{\rm day}) = & {\rm C}{\rm a}_{{\rm int}}({\rm g}/{\rm day}) - [{\rm C}{\rm a}_{{\rm fc}}({\rm g}/{\rm day}) \cr & - {\rm C}{\rm a}_{{\rm fce}}({\rm g}/{\rm day})]}$$where Caint is the Ca intake (g/day).
The biological availability of Ca was determined by dividing the Caabs (g/day) by the Caint (g/day), according to the equation:
$${\rm Biological}\,{\rm availability}\,{\rm of}\,{\rm Ca} = \displaystyle{{{\rm C}{\rm a}_{{\rm abs}}({\rm g}/{\rm day})} \over {{\rm C}{\rm a}_{{\rm int}}({\rm g}/{\rm day})}}\; $$The retained Ca (Caret) was determined using the following equation:
$$\eqalign{{\rm C}{\rm a}_{{\rm ret}}({\rm g/day}) = &[{\rm C}{\rm a}_{{\rm int}}({\rm g/day})] - [({\rm C}{\rm a}_{{\rm fc}}({\rm g/day}) \cr & + {\rm C}{\rm a}_{{\rm ur}}({\rm g/day})}]$$where Caur is the Ca in urine (g/day).
The daily net Ca requirements for maintenance were calculated from the intercept of the linear regression between daily excretion of Ca (i.e. sum of Cafc and Caur (g/kg BW)) against daily Caint (g/kg BW) using the minimum endogenous losses method (MEL).
Statistical analyses
The animals were assigned to one of three dietary treatments (0.6, 1.7 and 3.0 g Ca/kg DM) in a completely randomized design with six replications for each treatment. The statistical model used to analyse the data was:
$$Y_{ij} = \mu + t_i + e_{ij}$$where Y ij is the observed value in the portion that received treatment i in replication j, μ is the overall mean of the experiment, t i is the effect of treatment i that was applied on a portion and e ij is the error term for i treatment and j replication.
The effect of different Ca content in the diets on DM intake (DMI), Caint, Cafc, Caur, Cafce, Caabs, biological availability of Ca, Caret and Capl, recorded per goat, were analysed using the MIXED procedure of SAS (version 9.4, SAS Inst. Inc. Cary, NC, USA). The statistical model included treatment (i.e. Ca dietary level) as fixed effect and the error as random effect. Orthogonal contrasts were performed to compare treatment means, using the CONTRAST statement and orthogonal polynomial coefficients in the MIXED procedure.
Regression equations to predict Caabs, Caret, Caur and excretion of Ca using Caint as the regressor, to predict Cafce using DMI as regressor and to predict net Ca requirements for maintenance, were obtained as mixed models considering the treatment as fixed effect and the error as random effect. Linear and quadratic regressions were tested using the MIXED procedure, selecting the most adequate model based on significance of P-value and lowest error. Slopes and intercepts of each equation were estimated using the ESTIMATE statement of the MIXED procedure. Significance level was declared at P < 0.05.
Results
As Ca content in the diet increased from 0.6 to 3.0 g/kg, Caint, Cafc, Caur, Cafce and Caabs showed linear increases of 309, 154, 77, 68 and 342%, respectively (P < 0.05; Table 2). Conversely, Caur as a percentage of Caint decreased from 4.1 to 1.8% as dietary Ca content increased. In addition, the Cafc as a percentage of Caint decreased from 80 to 50% as Ca content in the diet increased from 0.6 to 3.0 g/kg DM, indicating that Ca is excreted mainly via faeces (Table 2).
Mean values of variables related to daily calcium (Ca) dynamics in Saanen goats fed diets containing different content of Ca
a Diets: 0.6 = 0.6 g Ca/kg DM; 1.7 = 1.7 g Ca/kg DM; and 3.0 = 3.0 g Ca/kg DM.
b SEM = standard error of mean.
c P-value refers to differences in parameters between different diets.
d Contrast: Orthogonal contrast: L = Linear effect; Q = Quadratic effect.
Dry matter intake reduced at an increasing rate (P < 0.01; Table 2) and Caret increased at a decreasing rate (P < 0.05; Table 2) with rising dietary Ca content. In contrast, Ca content in the diet did not affect BW (25 ± 4.6 kg), biological availability of Ca (0.66 ± 0.026) from diets or Capl (17 ± 4.6 mg/dl). The biological availability of Ca from limestone in Saanen goats was 0.72 ± 0.026 (Fig. 1; P < 0.01).
Relationship between daily calcium (Ca) intake (mg/kg body weight (BW)) and daily true absorbed Ca (mg/kg BW) for Saanen goats fed diets containing different content of Ca. RMSE = root mean square of error.
It was found that Caabs, Caret and Caur were associated linearly with Caint, which allowed the development of regression equations to predict Caabs (root mean square of error (RMSE) = 3.23; P < 0.01; Fig. 1), Caret (Eqn (11)), Caur (Eqn (12)) and Cafc (Eqn (13)) as a function of Caint:
$$\eqalign{&{\rm C}{\rm a}_{{\rm ret}} = - 12\,(\pm 1.6{\rm )} + 0 . 61\, (\pm 0 . 025{\rm )} \times {\rm C}{\rm a}_{{\rm int}} \cr &{\rm (RMSE} = 2 . 96,\,P \lt 0 . 01)}$$
$$\eqalign{&{\rm C}{\rm a}_{{\rm ur}} = 0 . 8\,(\! \pm 0 . 24{\rm )} + 0 . 01\,(\! \pm 0 . 004{\rm )} \times {\rm C}{\rm a}_{{\rm int}} \cr & {\rm (RMSE} = 0 . 39,\,P = 0 . 022)}$$
$$\eqalign{&{\rm C}{\rm a}_{{\rm fc}} = 10\, (\! \pm 1 . 3{\rm )} + 0 . 39\, (\! \pm 0.022{\rm )} \times {\rm C}{\rm a}_{{\rm int}} \cr & {\rm (RMSE} = 2 . 52,\,P \lt 0 . 01)}$$Additionally, it was found that Cafce was associated linearly with DMI (RMSE = 1.51; P < 0.01; Fig. 2). The daily Ca requirement for maintenance, estimated from the intercept of the linear regression between excretion of Ca and Caint, was 12 (±1.3) mg/kg BW (RMSE = 2.44; P < 0.01; Fig. 3)
Relationship between daily dry matter intake (g/kg body weight (BW)) and daily faecal endogenous calcium (Ca) (mg/kg BW), for Saanen goats fed diets containing different content of Ca. RMSE = root mean square of error.
Relationship between daily calcium (Ca) intake (mg/kg body weight (BW)) and daily excretion of Ca (mg/kg BW), for Saanen goats fed diets containing different content of Ca. RMSE = root mean square of error.
Discussion
The goals of the present study were to evaluate the dynamics of body Ca in goats using different content of Ca in diets and estimate the net Ca requirements for the maintenance of Saanen goats, using the isotopic dilution method.
Increasing Ca content in diets from 0.6 to 3.0 g/kg DM resulted in a decrease of the ratio between Cafc and Caint from 0.79 to 0.49, indicating that within the range of Ca content studied, Saanen goats improved their Ca use as the content of Ca in the diet increased, which is in agreement with the increase of Caabs observed with the increase of Ca content in diets. Since Ca availability did not change with the increase of Ca content in diets, the amount of true Ca absorbed was directly proportional to the dietary Ca content. In addition, because the P content in the diet did not change across treatments, the increase in Caabs with increase of Ca content was associated with an increase in the Ca:P ratio of diets from 0.24 in the low Ca diet (0.6 g Ca/kg DM) to 1.2 in the high Ca diet (3.0 g Ca/kg DM). This is related to an antagonistic relationship between Ca and P during Ca absorption processes in the intestine (Field et al., Reference Field, Suttle and Nisbet1975; Schneider et al., Reference Schneider1985). Phosphate, which is the principal dietary form of P, binds more easily to Ca and decreases its intestinal availability (Favus et al., Reference Favus, Bushinsky, Lemann and Rosen2009). Conversely, mechanisms of Ca and P absorption are mediated by both transcellular and passive cellular processes, with P absorption being more efficient than Ca absorption (Suttle, Reference Suttle2010). Active absorption of both minerals is regulated by 1,25-dihydroxyvitamin D (1,25(OH)2D) metabolism. When Ca requirements are fulfilled, if P concentration is greater than Ca concentration in the diet, 1,25(OH)2D promotes P absorption, which decreases Ca absorption in the intestine (DiMeglio and Imel, Reference DiMeglio, Imel, Burr and Allen2013). Thus, Ca absorption is regulated not only by its concentration but also by its ratio with P concentration in the diet. In addition, the ratio between Caret and Caabs increased from 0.21 to 0.72 as Ca content in the diets increased from 0.6 to 3.0 g/kg DM, indicating that a greater amount of Ca was incorporated into tissues with greater dietary Ca content and increased Caint. The Saanen goats used in the current study were in the growing phase, when Ca retention is directed principally to bone deposition (Suttle, Reference Suttle2010). Therefore, the evaluation of higher Ca contents in the diet of growing Saanen goats is recommended for future studies to estimate the maximum Caret achievable.
Although Caret and Caabs increased with increasing Ca in the diet, the biological availability of Ca and Capl were not affected by Ca content in the diets. The restriction of dietary Ca or excessive Ca supply is nutritional conditions that may alter the Capl and its biological availability in goats (NRC 2007; Suttle, Reference Suttle2010). In addition, the biological availability of Ca and Capl are principally associated with animal species, age, nutritional requirements, the chemical form of Ca in feed and dietary P concentration (Suttle, Reference Suttle2010; Wilkens et al., Reference Wilkens, Breves and Schröder2014). Therefore, it is suggested that since the Ca content of diets in the current study still met maintenance Ca requirements, the Capl and Ca biological availability did not change with the treatments and the amount that exceeded the maintenance requirements was designated to Ca body retention, which was clearly observed in the current study.
Values for biological availability of Ca between 0.30 and 0.55 and for Capl between 8 and 12 mg/dl have been accepted by the NRC (Reference Wilkens, Breves and Schröder2007) for goats. Despite this, the NRC (2007) does not discriminate between the values of biological availability of Ca and the Capl between different ruminant species. In the current study, the average values for biological availability of Ca from diets and Capl were 0.66 and 17.3 mg/dl, respectively, which were greater than those reported by the NRC (2007). This difference might be due to the mineral concentrations and metabolism in tissues, blood and milk being different between ruminant species (Haenlein, Reference Haenlein1980; Herm et al., Reference Herm2015). In addition, the biological availability of Ca from limestone in Saanen goats (0.72) was greater than reported in lambs (0.62; Roque et al., Reference Roque2007). As demonstrated by Wilkens et al. (Reference Wilkens2012, Reference Wilkens, Breves and Schröder2014), the Ca absorption and active transport in jejunum processes are more efficient in goats than sheep. Herm et al. (Reference Herm2015) suggested subsequently that differences in the expression of sodium/Ca exchanger type 1, vitamin receptors and 24-hydrolase may explain the difference in Ca absorption between sheep and goats. Therefore, differences between the findings of the current study and those in the literature highlight the importance to identify the mineral's biological availability for each ruminant species and further studies are needed.
Faecal endogenous Ca increased as Ca content in the diet increased. Additional studies evaluating the relationship between faecal endogenous Ca and its content in the diet have not been conducted in goats, which makes comparison of the current results with those in the literature difficult. However, Carvalho et al. (Reference Carvalho2003), using the isotopic dilution method, found that castrated Saanen goats of 25 kg BW showed increased faecal endogenous P as P content in the diet increased. This evidence is in accordance with the current study, considering that Ca and P are metabolically related. Conversely, there is a consensus in the literature that Cafce only increases with higher DMI but not with an increase of Caint (Braithwaite, Reference Braithwaite1982; AFRC 1998; NRC 2007). However, the current study does not support those findings. This physiological response was associated with the negative correlation between DMI and dietary Ca content. In comparison to other ruminant species, when subjected to dietary Ca restriction, goats prioritize active Ca absorption over bone Ca resorption (Wilkens et al., Reference Wilkens, Breves and Schröder2014). Considering that goats in the current study decreased DMI with increasing of dietary Ca and decreased Cafce with increasing DMI, goats possibly prioritize Ca absorption over Ca intake.
The current study is the first to evaluate endogenous Ca responses to different Ca content intakes in goats using isotopic dilution, the majority of previous studies related to this subject in the literature were performed with sheep and cows. Therefore, it is possible that there are differences in Ca digestion and metabolism between ruminant species and more studies with goats are needed to better elucidate these mechanisms for this species. This will be important for enhancing diet formulation for goats, because the majority of Ca recommendations for goat feeding systems (AFRC 1998; NRC 2007) are based mainly on DM intake.
The current result for daily Ca requirement for maintenance was lower than that reported for goats by the NRC (2007) using feeding trials (20 mg Ca/kg BW), whereas it was closer to that reported for sheep by the ARC (1980) using the isotopic dilution method (15.7 mg Ca/kg BW). Comparative slaughter studies with male crossbreed Boer × Saanen (Fernandes et al., Reference Fernandes2012; Teixeira et al., Reference Teixeira2015) and with Saanen goats of different sexes (i.e. castrated males, females and intact males) (Santos et al., Reference Santos2016) have also reported greater values of daily Ca requirements for maintenance (27.4, 38.3 and 35.4 mg Ca/kg BW, respectively) than those found in the current study.
The differences between the current results and the reports mentioned previously, especially the study using Saanen breed at a similar BW (Santos et al., Reference Santos2016), may be related to the method used. Comparative slaughter estimates apparent Ca inputs and outputs, in comparison to the isotopic dilution method that is able to obtain true estimates of Ca inputs and outputs (Vitti and Kebreab, Reference Vitti and Kebreab2010). Thus, additional studies should be conducted to improve the accuracy of the estimation of Ca requirements for the maintenance of goats trying to cover more different stages of growth.
The current study seems to be the first to evaluate the dynamics of Ca and Ca requirements for maintenance with different content of Ca in diets in goats using the isotopic dilution method. The current findings might help understand Ca dynamics in goats and promote the formulation of diets adequately balanced in terms of Ca for growing Saanen goats. In conclusion, the ratio between retained Ca and true absorbed Ca increased as Ca content in the diets increased from 0.6 to 3.0 g/kg DM, suggesting that additional studies to evaluate Ca content in diets greater than 3.0 g/kg DM are required to estimate the maximum retained Ca in growing Saanen goats. It was also concluded that Ca true absorption and plasma Ca remain similar if dietary Ca supply is above of maintenance requirements of Ca. In addition, daily Ca requirements for maintenance was 12 (±1.3) mg/kg BW.
Financial Support
The authors thank the São Paulo Research Foundation (FAPESP) for providing financial support (grants # 05107-0/1996 and 58351-5/2008).
Declaration of interest
Authors declared no conflicts of interest.
Ethical standards
Humane animal care and handling procedures were conducted in accordance with the Animal Care Committee (Comissão de Ética e Bem Estar Animal) of the Sao Paulo State University and with instructions from the Ministry of Agriculture in Brazil (instruction number 56/2008).
