| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
,2
* Animal Nutrition Laboratory, Centro de Energia Nuclear na Agricultura, Caixa Postal 96, CEP 13400-970, Piracicaba, São Paulo, Brazil;
and
Centre for Nutrition Modelling, Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario N1G 2W1, Canada
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
and extend its use to study Ca flows in growing sheep. Twelve Santa Ines male sheep, 8 mo of age, with average BW of 31.6 kg were injected with 32P and 45Ca to trace the movement of P and Ca in the body. The original model had 4 pools representing the gut, plasma, soft tissues, and bone. In the revised model, instantaneous values rather than averages for pool derivatives were incorporated, and the model was extended to represent absorption and excretion of phytate P explicitly. The amendments improved the model, resulting in higher flows between plasma and bone than between plasma and tissue and, therefore, a more accurate representation of P metabolism. Phosphorus and Ca metabolism were then assessed conjointly using the revised model. The results showed that P and Ca metabolism are closely related as evidenced by the ratio of these minerals in the bidirectional flows between plasma and bone and between plasma and tissue. Phytate P digestibility was 47%, and P retention was negative (–1.4 g/d), suggesting that a feed characteristic impaired P utilization and led to P deficiency. The revised model provides an improved prediction of P and Ca metabolism that can be used to assess mineral requirements and to estimate losses to the environment.
Key Words: calcium • kinetics • modeling • phosphorus • sheep
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
). Moreover, excess P excreted in feces represents an unnecessary cost to farmers. Knowledge about the mineral content of feedstuffs and P quality, as well as understanding of P kinetics, is beginning to affect P supplementation (Bravo et al., 2003).
Calcium can influence P utilization because of their close metabolic relationship (Tamim and Angel, 2003). These minerals are present mainly in bone as hydroxyapatite salt and are mobilized when an animal’s requirements are not met by the diet. Chapuis-Lardy et al. (2004) suggest that limestone used as a Ca supplement in the diet can affect fecal excretion of Ca and P because of Ca-P complex formation along the digestive tract reducing the bioavailability of both minerals.
Diets based on cereal grains are used to promote rapid growth of ruminants. However, incorporation of cereals in diets leads to an increase in phytate P content, making it necessary to evaluate phytate digestibility by ruminants fed cereal-based diets. Phytate P is hydrolyzed by phytase produced by microbes in the rumen; nevertheless, enzyme efficiency can be altered by different factors such as feed treatment (Bravo et al., 2002), presence of Ca and Mg (Dutton and Fontenot, 1967), rumen pH, and amount of phytate P in the diet (Ellis and Tillman, 1961).
The use of isotope dilution techniques provides information on endogenous P, which represents a large part of P excreted in feces. Moreover, it is possible to study P kinetics and improve models of P metabolism by using these techniques. A 4-pool model for studying P kinetics was proposed by Vitti et al. (2000) to assess metabolism of this mineral in growing goats. Metabolism of P and Ca are related and similar.
The aim of this study was to extend the Vitti model and apply it to study P and Ca flows in growing sheep, thereby providing a better understanding of the metabolism of these minerals in ruminants.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
). Thus, there were 4 treatments with differing Ca sources, and 3 animals were fed each treatment. The diets were formulated to meet nutrient requirements of the sheep according to NRC (1985) and to provide the same Ca and P levels. Feed was offered twice daily at 0800 and 1700.
|
Analysis of Samples.
Samples of feed and feed refusals were analyzed for DM, CP, P, Ca, NDF, and ADF (Table 2). Crude protein was determined by the Kjeldahl method, ADF following AOAC (1995), and NDF according to Mertens (2002) without amylase or sulfite. The samples were analyzed for phytate P according to Latta and Eskin (1980).
|
Phosphorus content was determined by colorimetry (Sarruge and Haag, 1974), and Ca was determined by atomic absorption spectrometry (Zagatto et al., 1979). Blood samples were centrifuged at 1,100 x g for 10 min, and the plasma was separated. After protein precipitation (1 mL of plasma and 9 mL of 100 g/L of trichloroacetic acid), inorganic P was determined by the method of Fiske and Subbarow (1925). Daily urine samples (30 mL) were collected and acidified with 12 M HCl to prevent volatilization of ammonia N and were thereafter dried (55°C) and ashed (500°C). Fecal samples were dried at 105°C, ashed at 500°C, and the ash was dissolved in 5 mL of 12 M HCl. Total P was determined using vanadatemolybdate reagents in a colorimetric method (Sarruge and Haag, 1974).
Radioactivity was measured by liquid scintillation counting and was corrected for decay and quench (Horrocks and Peng, 1971) by the external standard technique (external source channels ratio method; Nascimento Filho, 1977). Feces samples (1 g) were dried (105°C) and ashed (500°C), and then were dissolved in 9M H2SO4 for 1 h. Samples of plasma (1 mL), urine (1 mL), and solubilized feces (1 mL) were counted in 10 mL of a scintillation solution using a Beckman Liquid Scintillation Spectrometer (model ls 5000 TA, Beckman-Coulter Inc., Fullerton, CA).
Revised Model
The amended Vitti model is shown in Figure 1. Like the original, it contains 4 P pools: 1) gut lumen, 2) plasma, 3) bone, and 4) soft tissue. The flows of P between pools and into and out of the system are shown as arrowed lines. The gut lumen, bone, and soft tissue pools interchange in 2 directions with the plasma pool, with flows F21 and F12, F23 and F32, and F24 and F42, respectively. Entry of P into the system is via intake, F10, and exit is via feces, F01, and urine, F02. Further, F10 is partitioned into P of phytate and nonphytate origins, and F01 and F21 are partitioned into P of dietary phytate, recycled endogenous, and dietary nonphytate origins (Figure 1).
|
After 7 d, pool 1 (gut lumen) was assumed to be in complete steady state (i.e., isotopic and nonisotopic steady state) and pool 2 (plasma) in nonisotopic steady state. The algebraic solution to the model (following Vitti et al., 2000; see Table 3 for principal mathematical notation) then gives:
|
 | [1] |
 | [2] |
 | [3] |
 | [4] |
 | [5] |
 | [6] |
 | [7] |
where the tilde indicates an experimentally determined flow, the specific activity of the pool formed by combining pools 3 and 4 is
 | [8] |
and the rate of change of specific activity of pool 2 is
 | [9] |
The values of s2 and R2 at 7 d were obtained after fitting the equation, s2(t) = ae–bt, to the plasma specific activity decay curve, where t (d) is time since administration of dose, and a (dpm/g) and b (per d) are parameters of the exponential equation. Differentiation of this equation with respect to time gives the instantaneous rate of change, 
, which permits calculation of R2 at 7 d.
The equations for calculating F32 and F42 (Eq. 3 and 4, above) differ from those presented by Vitti et al. (2000; Eq. 20 and 21); in solving the original Vitti model, average rather than instantaneous values for ds3/dt and ds4/dt were used to derive their Eq. [20] and [21]. Instantaneous values for ds3/dt and ds4/dt were used here to derive Eq. [3] and [4], which were obtained by assuming the relative rates of change of specific activity of pools ds2 1 ds3 2, 3, and 4 are the same after 7 d; i.e., 
If ingested phytate P, 
10(p), is measured, then P excretion in feces, F01, and P absorption, F21, can be partitioned between P sources using the following equations:
 | [10] |
 | [11] |
 | [12] |
 | [13] |
 | [14] |
 | [15] |
where
 | [16] |
 | [17] |
Superscript (p) denotes P of dietary phytate origin, (e) recycled endogenous P, and (n) P of dietary nonphytate origin. The model is solved by computing Eq. [1] through [17]. The same basic scheme (Figure 1
) applies to Ca, and in this case the model is solved by computing Eq. [1] through [9] only.
Statistical Analyses
Experimental measurements of P intake and excretion, total P in bone and tissue, and specific activity were analyzed as a completely randomized design, as were the flows calculated from the model. Variables were subjected to ANOVA using the GLM Procedure of SAS (SAS Inst. Inc., Cary, NC), and the LSMEANS procedure was used to calculate means and standard errors. Regression analysis was carried out using the PROC REG procedure.
A comparison of observed and predicted values of specific activity in plasma for Ca and P, and of phytate (F01(p)) and other P (F01(e) + F01(n)) excreted in feces was made using the mean square prediction error (MSPE), calculated as
 |
where n is the number of observations, Oi is the observed value, and Pi is the predicted value. Root MSPE expressed as a proportion of the observed mean gives an estimate of the overall prediction error. The MSPE was decomposed into error due to random variation (ED), error due to deviation of the regression slope from unity (ER), and error due to central (mean) bias (Bibby and Toutenburg, 1977).
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
) was adequate for growing sheep according to NRC (1985). Total P excreted in feces (F01; Table 4) was greater than intake leading to negative P retention (–1.40 g/d). Total P excreted in feces was separated into dietary phytate P (F01(p)), dietary nonphytate P, (F01(n)) and endogenous P (F01(e)) (Figure 1). The measured and predicted values of F01(p) and F01(n) + F01(e) are shown in Table 5. Endogenous fecal P was 2.50 g/d, fecal P of dietary phytate origin was 0.68 g/d, and fecal P of dietary nonphytate origin was 1.29 g/d as measured, and these fractions represent 56, 15, and 29% of F01, respectively. Total P excreted in feces was positively related to P ingested, where
|
|
 |
Phosphorus absorbed (F21) was also separated into the same 3 fractions. The mean values for F21(p) and F21(n) + F21(e) were 0.49 and 2.59 g/d, respectively.
Specific activities and P content in urine, plasma, bone, and soft tissue are also summarized in Table 4. Predicted and measured values of specific activity in plasma P at 144 h are shown in Table 5. Flows of P between pools are summarized in Table 6.
|
. Ingestion of Ca satisfied requirements for growing sheep established by NRC (1985). Mean measured Ca retention was 0.09 g/d and was related to Ca absorption (G21) by
 |
Total Ca excreted in feces increased directly with Ca intake as given by
 |
Specific activities and Ca contents are summarized in Table 4. Predicted and measured values of specific activity in plasma Ca at 144 h are shown in Table 5.
Calcium and Phosphorus
The amounts of P and Ca ingested (F10 and G10) were the same for all diets, justifying assemblage of the data into a single, combined data set. Diets were representative of 4 treatments, each supplemented with a different Ca source, namely citrus pulp, limestone, alfalfa hay, and oyster shell meal. Data were analyzed to determine if different sources of Ca affected Ca and P kinetics, but no significant effect was found (data are not presented).
The Ca:P ratio present in bone was 1.97 and that present in tissue was 0.17. The Ca:P ratio for mobilization from bone (G23/F23) was 2.50 and that for accretion to bone (G32/F32) was 2.58. These ratios were correlated as
 |
The Ca:P ratio for mobilization from tissue to plasma (G24/F24) was 0.20 and for accretion in tissue (G42/F42) 0.23, and both were correlated:
 |
Calcium excreted in feces increased with P excretion in feces according to
 |
Flows of Ca between pools are summarized in Table 6.
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
; Viveros et al., 2000) and soybean meal (Morse et al., 1992; Ravindran et al., 1994). For alfalfa hay, phytate P was greater than reported by Morse et al. (1992), though it compared closely with that given by Ravindran et al. (1994) for leaf meals. Differences in phytate P in alfalfa hay could be related to factors such as the age of the plant, cultivars, processing conditions, height of cutting, number of observations per feedstuff, as well as analytical methods used (Eeckhout and De Paepe, 1994).
Phytate P digested (intake minus feces, as measured) was 0.61 g/d, with a digestibility of 47%. This result is lower than some reported in the literature (Skrivanova et al., 2004; Kebreab et al., 2005).
Predicted values for F01(p) and F01(e) + F01(n) were related to measured values as shown by the MSPE analyses in Table 5. The ED percentages were 60% for F01(p) and 85% for F01(e) + F01(n). For both predictions, the bulk of MSPE was due to random variation in the data set. Phosphorus utilization has a large variation among animals (Field et al., 1983; Rapp et al., 2001), so it is reasonable to assert that the ED percentages indicate satisfactory prediction. There was slight overprediction for F01(p), and slight underprediction for F01(e) + F01(n) with the mean bias contributing about 13% of MSPE for both. The ER were 26 and 0% of total MSPE for F01(p) and F01(e) + F01(n), respectively.
Flows F21 and G21 represent total P and Ca flowing from gut to plasma, and the flow was higher for P (Table 6). However, P and Ca of dietary origin digested were 1.22 and 1.48 g/d as measured, giving digestibilities of 39 and 30%, respectively, for P and Ca. The low values for P absorption and digestibility are related to low phytate digestibility, which likely influenced Ca absorption and its availability as well.
Phytate P digestibility was higher than dietary non-phytate P digestibility. This may have happened because phytate P can be hydrolyzed in the large intestine, leading to overestimation of phytate P absorption, as this hydrolyzed phytate P is measured as inorganic P in feces (Godoy and Meschy, 2001; Matsui, 2002; Shen et al., 2005). Moreover, phytate P degraded in the large intestine is unlikely to have nutritional significance for ruminants (Park et al., 2002) because the absorption of P occurs in the small intestine (Pfeffer et al., 1970; Grace et al., 1974; Care, 1994). At the same time, large intestinal hydrolysis of phytate P would lead to underestimation of the digestion of nonphytate P of dietary origin.
Predicted values for specific activities in plasma P and Ca 144 h after radioisotope injection were very close to observed values as shown by MSPE analysis. Percentages of total MSPE as error due to central (mean) bias, ER, and ED were 3, 18 and 79%, respectively, for P and 4, 21 and 74%, respectively, for Ca. This lends confidence to the projected values for instantaneous specific activities in plasma P and Ca at 168 h after radioisotope injection.
The use of instantaneous values for pool derivatives based on projected specific activities in plasma, rather than the use of average values for derivatives, improves reliability of the model of Vitti et al. (2000). This claim is supported by the greater flows between plasma and bone than between tissue and plasma (Table 6). It is well known that 80 to 85% of total P present in an animal’s body is present in bone, which represents an important reserve of Ca and P that can be mobilized for animal function. On the other hand, P present in soft tissue can also be mobilized, but in lower proportions. Thus, the revised model more accurately reflects the higher values of P and Ca deposition in and resorption from bone.
Although the animals ingested the required amounts of Ca and P, high excretion of P (F01) and Ca (G01) led to low retention of both minerals. The relationship between F01 and G01 suggests formation of a Ca-P complex in feces, which agrees with the observations of Chapuis-Lardy et al. (2004), who found that fecal Ca has a major impact on the proportion of water soluble P in total P excreted in feces of dairy cows.
Phosphorus mobilized from bone (F23) was greater than P accreted (F32; Table 6). These data are in agreement with Vitti et al. (2000) who, in studying metabolism in growing goats fed different levels of P, suggested that mobilization of P from bone occurs to maintain metabolism in growing animals, which require high amounts of P for their development. However, values for tissue resorption (F24) greater than for absorption to tissue (F42) suggest that, as well as mobilizing P from bone, the animals were mobilizing P from tissue to satisfy requirements. Low P digestion likely impaired utilization of dietary P, leading to P deficiency. According to Valk et al. (2002), dairy cows mobilize P from body reserves to compensate for excretion of P in milk and feces.
The ratio of Ca:P in bone was 1.97, but the ratios of these minerals being mobilized and accreted in bone were 2.49 and 2.59 g/d, respectively. The ratio in bone approaches the stoichiometry of hydroxyapatite (Ca:P mass ratio 2.15:1.0), the source of mobile P in bone for regulating plasma P. Higher ratios of Ca:P for mobilization and accretion than in hydroxyapatite suggest that, because G32 was slightly greater than G23, the bone mineral could be other than hydroxyapatite. Shapiro and Heaney (2003) reported that rats supplemented with Ca showed higher a Ca:P ratio in their bone, possibly due to amorphous crystals with a large component of calcium carbonate.
Young growing sheep were used in this study, and it is assumed that these animals have good ability to adapt to Ca and P deficiency and that bone resorption and formation can occur simultaneously.
The relationships obtained between Ca:P ratio mobilized and accreted in bone and soft tissue support the claim that these minerals are very closely related, and therefore, it can be assumed that impairment of P digestibility resulted in a lower Ca digestibility. However, the mechanisms involved are not clear, and more research is needed to assess this issue.
Serum P was greater than normal. As serum P is influenced by P mobilization from bone and animals deplete their bone reserves to support their metabolic functions (Wu et al., 2001), P likely was mobilized from bone to maintain serum P levels. It can also be speculated that high values of plasma P led to high salivary P concentrations. According to Valk et al. (2002), there is a direct positive relationship between plasma P and salivary P concentrations. Because endogenous P in feces (F01(e)) comes mainly from salivary secretion, it can be argued that P secretion through saliva reflects an attempt to maintain P metabolism, leading to a high proportion of endogenous P in total P excreted in feces (F01(e)/F01). The absorption of other P (F21(n) + F21(e)) being higher than absorption of phytate P (F21(p)) supports this argument.
Excretion of P through urine is considered negligible in ruminants, and it is related to efficiency of absorption and to plasma P concentrations relative to the renal threshold (6.5 to 9.5 mg/dL; Challa and Braithwaite 1988; Challa et al., 1989). In our study, plasma P was higher than normal (9.3 mg/dL) despite negative P retention. Therefore, it is likely that this favored excretion of P in urine (F02).
|
IMPLICATIONS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
|
Footnotes |
|---|
2 Corresponding author: ekebreab{at}uoguelph.ca
Received for publication February 14, 2006. Accepted for publication May 31, 2006.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONIMPLICATIONSLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. R. Hill, K. F. Knowlton, E. Kebreab, J. France, and M. D. Hanigan A Model of Phosphorus Digestion and Metabolism in the Lactating Dairy Cow J Dairy Sci, May 1, 2008; 91(5): 2021 - 2032. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
