| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Dairy Science, Virginia Polytechnic Institute and State University, Blacksburg 24061
Virginia Maryland Regional School of Veterinary Medicine, Blacksburg 24061
University of Pennsylvania, Kennett Square 19348
1 Corresponding author: knowlton{at}vt.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: bone • calcium • phosphorus • cow
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Approximately 98% of total body Ca and 80% of body P are located in the skeleton. Bone Ca and P are stored as calcium phosphate [Ca2(PO4)2], which is amorphous, and as hydroxyapatite [Ca10(PO4)6(OH)2], which is a crystalline structure that provides binding locations for Ca and P. Mobilization of bone hydroxyapatite results in 10 ions of Ca and 6 ions of P being released into circulation (Goff, 2000) and is driven primarily by the concentration of blood Ca rather than blood P. The timing of bone resorption and formation are currently unknown in the dairy cow, but several blood markers of bone metabolism are available, allowing bone formation and resorption to be monitored noninvasively. However, these markers have yet to be validated in dairy cows with bone samples and mineral balance.
By accounting for drafts on bone with different dietary Ca concentrations, the P concentration of the diet could potentially be reduced to coincide with endogenous P release from bone during times of Ca-induced bone resorption. The amount of P excreted in the feces of dairy cows is directly proportional to excess dietary P (Morse et al., 1992). Phosphorus run-off is one of the leading causes of fresh water eutrophication and is a major focus of nutrient management for livestock producers (Knowlton et al., 2004). Field surveys demonstrate that P is typically fed at 20 to 40% above published recommendations for dairy cows (Knowlton et al., 2004). If these recommendations are excessive because of the lack of information on bone mineral resorption, the overfeeding problem is even worse than currently surmised. A more detailed understanding of bone mineral metabolism could help decrease excretion of excess P.
Our goal was to evaluate changes in body Ca and P during lactation based on Ca and P balance, serum markers of bone metabolism, and bone biopsy samples in cows fed 1 of 3 dietary concentrations of Ca. Our hypothesis was that feeding a low Ca diet will cause an increase in bone mobilization in early lactation that will result in an increase in the endogenous pool of P.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
). Dietary energy and protein were maintained by substitution of corn, soybean meal, and tallow for soy hulls and limestone. Dietary P concentration was 0.34% in all diets. Before calving, all cows consumed the same dry cow diet (Ca 0.39% and P 0.38%). Treatment groups were balanced for parity and mature equivalent milk production. Treatments were applied from calving through 140 d of lactation. When cows were not in the metabolism stalls for total collection (described below) they were group-housed in a free stall barn and fed via a Calan gate system (American Calan, Northwood, NH). Cows were fed ad libitum once daily at 1530 h, feed offered and refused was recorded daily, and feed offered was increased daily if there were less than 10% refusals. Cows were milked twice daily at 0300 and 1500 h, and milk yield was recorded at each milking throughout the study. All procedures and protocols were approved by the Virginia Tech Animal Care and Use Committee.
|
Cows were milked twice a day at 0530 and 1730 h while standing in the metabolism stalls. Milk was weighed and sampled at all milkings, and samples were analyzed for fat, protein, SCC, SNF, and lactose (Dairy Herd Improvement Association, Blacksburg, VA). A second milk sample was stored frozen at –20°C and later analyzed for Ca and P by the method of Walter et al. (1997; CEM Corporation, Matthews, NC).
In the metabolism barn, cows were fed at 0600 h; daily feed offered and refused was recorded during the balance periods. Feed refusals for each cow were sampled on d 3 of each balance period. Individual ration ingredients were sampled weekly throughout the study and pooled by month. Feed and feed refusal samples were dried at 60°C to a constant weight, ground through a 1-mm screen in a Wiley mill (Arthur H. Thomas), and analyzed for DM, NDF, ADF, Ca, and P concentrations by Cumberland Valley Analytical Services in accordance with approved AOAC methods.
Blood Sampling and Analysis
Jugular blood samples were obtained at 14 and 10 d (averaged -12 and -8 d) before expected calving, at calving, and at 1, 3, 5, 10, 14, 21, 28, 35, 42, 56, 70, 84, 98, 119, and 140 d postcalving. Samples were collected and immediately placed on ice until centrifugation (2,200 x g for 20 min). Serum separators tubes (Fisher Scientific, Pittsburgh, PA) were utilized to facilitate serum separation during centrifugation. Serum was harvested and stored frozen at –20°C until analysis. Colorimetric methods were used to analyze all serum samples for Ca (Aresenazo Reagent Set, Pointe Scientific, Canton, MI) and inorganic P (Inorganic Phosphorus Reagent Set, Pointe Scientific, Canton, MI). The interassay and intraassay coefficients of variation for serum Ca and serum P were 3.6 and 3.4%, respectively. Competitive immunoassays were used to quantify serum osteocalcin (OC; meta osteocalcin, Quidel Corporation, San Diego, CA) and deoxypyridinoline (DPD; total deoxypyridinoline, Quidel Corporation) in all serum samples. Serum OC had an interassay coefficient of variation of 2.8% and intraassay coefficient of variation of 2.7%, whereas interassay and intraassay coefficients of variation were 2.8 and 2.7%, respectively, for serum DPD. Parathyroid hormone (PTH) was analyzed in a pooled precalving sample (–14 and –10 d) and in the 1, 3, 5, and 10 d samples by a radioimmunoassay specific for the intact PTH peptide (N-tact PTH SP, DiaSorin, Saluggia, Italy). The PTH assay was completed in one run with an intraassay coefficient of variation of 9.1%.
Bone Biopsies
Rib bone samples were collected on d 8 ± 3.5 (left side, 11th rib) postcalving and on d 77 ± 5.6 (right side, 11th rib) and d 140 ± 3.7 (left side, 12th rib) using a modification of the procedure of Beighle et al. (1993). Briefly, biopsies were performed with the animal restrained in a standing head gate. The predesignated side for the bone sample was clipped ventrally for 25 cm beginning below the lumbar vertebrae and 20 cm caudal and cranial to the sample rib. The clipped area was cleaned and surgically prepared with alternate application of Betadine Surgical Scrub (Purdue Pharma L. P., Stanford, CT) and alcohol. The incision site was anesthetized with 2% lidocaine in an inverted L pattern. After a second surgical scrub and alcohol rinse, a 10-cm incision was made over the midline of the rib through the skin, fascia, and muscle. The periosteum was displaced from the biopsy site with a periostal elevator (Jorgensen Laboratories, Loveland, CO), and the biopsy site was exposed with a Weitlaner retractor (Jorgensen Laboratories). A power drill and sterile drill bit were used to make a guide hole for the bone trephine (16 mm, Galt cranial trephine, Mercedes Medical, Tellevast, FL). Trephine depth was restricted to allow for sampling of the exposed medial and lateral bone to the depth of the medullary cavity. Bone samples weighed 700 to 1,000 mg. The circular piece of bone was removed and immediately stored in a sealed plastic bag and placed on ice. The periosteal flaps were replaced loosely without suturing followed by suturing of the muscle and stapling of the skin separately.
Bone samples were extracted with ether using a modified Soxhlet procedure (AOAC, 2006) and subsequently ashed at 600°C in a muffle furnace. Ash samples were then analyzed for Ca and P using a microwave digestion procedure. Briefly, ashed samples were digested in 15.7 N nitric acid in a MARS 5 microwave (CEM Corporation, Matthews, NC) at 190°C and 14.1 kg/cm2 for 10 min. Samples were further digested with 30% hydrogen peroxide for 30 min, diluted, and analyzed for Ca and P content colorimetrically.
Statistical Analysis
All data were analyzed using the MIXED procedure of SAS (2003, SAS Institute Inc., Cary, NC) with the model defined in Table 2. Blood analyses included the –14 and –10 d samples which allowed for prepartum representation on the graphs. Two first-lactation cows consuming the MED diet did not have wk-2 balance data because of illness; one cow was treated for pneumonia, and another cow had a displaced abomasum at 10 d postpartum. Data from these 2 cows were removed from the analysis during times of illness (d 10 to 18). The wk –2 collection data (cows were consuming the same prepartum diet) were used as covariates for the feces and urine variables with the exception of one HI cow that calved 2 wk before her expected due date. The –2 wk data collected from the other HI cows were averaged and used as the covariate for that cow.
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
).
|
). Multiparous cows consumed more DM than did primiparous cows (25.4 vs. 21.4 ± 0.50 kg/d) and there was a significant parity x week interaction. Dry matter intake responded both linearly (P < 0.0001) and quadratically (P < 0.0001) over time; however, the multiparous cows had a more rapid increase in DMI for the first 4 wk of lactation. Intake by primiparous cows peaked around wk 14, whereas multiparous cows peaked during wk 11. Dry matter intake was different across week, with cows consuming an average of 11 kg/d DM during wk 1, 27.4 kg/d at peak, and 27.2 kg/d DM at the conclusion of the study.
|
). As expected, multiparous cows had greater daily milk yield as compared with primiparous cows (36.3 vs. 29.5 ± 0.86 kg/d). Diet had no effect on milk fat yield, but there was a significant interaction of parity and time. Multiparous cows had greater milk fat yield because of greater production, with the exception of wk 8 and 20 in which the 2 parity groups were not different. There was a significant treatment x week and parity x week interaction for milk protein yield. Milk P content averaged 0.88 g/kg, and milk Ca content averaged 1.13 g/kg; treatment had no effect on these measures (data not shown).
Serum Minerals
Serum Ca was not affected by dietary Ca concentration (Table 5; Figure 1). At calving cows were within or just below the normal serum Ca range of 8 to 10 mg/dL (Goff, 2000) with no clinical signs of hypocalcemia. Similar to our findings, dietary Ca concentration had no effect on plasma Ca concentration in sheep fed 0.47 or 0.82% dietary Ca (Takagi and Block, 1991) or in postpartum cows fed 0.99 or 1.5% dietary Ca prepartum (Chan et al., 2006). There was a linear response in serum Ca over time (Table 5; Figure 1). First lactation cows tended to have greater serum Ca concentration than the multiparous group (P < 0.08; 11.45 vs. 10.76 ± 0.27 mg/dL). Chan et al. (2006) also observed that primiparous cows had greater serum Ca than multiparous cows.
|
|
). Chan et al. (2006) also reported no difference in postpartum serum P concentration when cows were fed dietary Ca at 0.99 or 1.50% of diet in the prepartum period. In the present study, serum P concentration was greater in first lactation cows compared with multiparous cows (5.7 vs. 4.9 ± 0.13 mg/dL). However, there was a significant 3-way interaction of treatment, parity, and day for serum P (P < 0.04) concentration, making main effects and 2-way interactions difficult to interpret on their own.
Serum Markers of Bone Metabolism
The interaction of treatment x parity was significant for serum OC concentration (Figure 2). Within the primiparous group, the MED cows had less OC concentration as compared with the other dietary treatments, whereas OC concentration was not affected by treatment in multiparous cows. Osteocalcin concentration reflects formation of the protein matrix; the biological explanation for this observation is not apparent.
|
; Figure 3A and B). These results are similar to a study by Kamiya et al. (2005) that examined the effect of prepartum dietary Ca concentration (0.46 or 0.86%) on bone turnover from 13 d prepartum to 3 d postpartum in primiparous and multiparous Holstein cows. In that study, there was no effect of dietary Ca concentration on plasma OC or urinary DPD, but primiparous cows had greater concentrations of both as compared with multiparous cows (Kamiya et al., 2005). This is likely because the younger cows were still growing and had a greater rate of bone turnover than the older cows.
|
) for serum DPD concentrations. Both parity groups had a negative linear response in DPD over time. Primiparous cows had greater (P < 0.05) serum DPD concentration than multiparous cows from –14 d prepartum to d 3, at d 10, and again at d 35 and 77. Otherwise the 2 groups were not different. Kamiya et al. (2005) found no effect of parity over time on urinary DPD concentration, but that study concluded at 3 d postpartum.
Both serum OC and DPD concentrations varied with time (Table 5). In early lactation serum OC was low in both primi- and multiparous cows suggesting relatively low formation of bone protein matrix. Osteocalcin then increased until 35 DIM where both groups reached a plateau. There was a linear decrease in serum DPD concentration over time (2.8 ng/mL at calving; 1.5 ng/mL at d 140), suggesting reduced bone formation immediately after calving. Liesegang et al. (2006) reported similar results for the marker of bone resorption they used, cross-linked carboxyterminal telopeptides of Type-I collagen (CTx), with an increase in serum concentration after parturition until 9 d postpartum. It was anticipated that serum DPD would be high around calving when the cows would likely be resorbing bone, but the lack of effect of dietary Ca was counter to our hypothesis.
Ekelund et al. (2006) found an interaction of treatment and time for plasma OC from cows consuming normal vs. low to normal dietary P concentrations (0.43 vs. 0.32% for the first 4 mo of lactation then 0.43% for the remainder). These authors concluded there was net bone formation in mid lactation (wk 17 to 24) as indicated by greater concentrations of OC during those weeks. However, there was not a corresponding decrease in concentration of CTx. Also, they saw no change in P retention as measured by total collection.
Peterson et al. (2005) observed no effect of prepartum dietary P on blood osteocalcin or deoxypyridinoline. Naito et al. (1990) observed that plasma osteocalcin was positively correlated with both plasma Ca and plasma inorganic P in periparturient cows. Those relationships were not observed in the current study. Plasma Ca was in the normal range prepartum, and whereas dietary Ca was less than is common in the field, Ca intake (~54 g/d during total collection 14 d prepartum) is in excess of the NRC requirement (35 to 45 g of Ca/d) and well above the intake thought to stimulate early bone resorption (10 g/d of Ca; Goings et al., 1974).
Bone is a dynamic tissue constantly being resorbed and formed; the relative rates of resorption and formation determine net bone mass. Parity clearly has an effect on bone metabolism, as does stage of lactation. When serum concentration of markers of formation and resorption over time are compared in the present study, the graphs appear to be mirror images (Figure 3 A and B). Serum OC concentration peaked around d 35 of lactation; simultaneously, DPD concentration began to decrease. The change in direction of serum DPD and OC may provide an indicator of a net change from bone resorption to bone formation, but closer evaluation suggests that this relationship only holds in cows fed adequate Ca. In cows fed the MED and HI diets, the correlation between OC and DPD was strong and negative (–0.41 and –0.71, respectively). In cows fed the LOW diet, however, there was no correlation between OC and DPD.
Serum Parathyroid Hormone
There was no effect of dietary Ca treatment on serum PTH concentration (9.9 ± 2.9 pg/mL). Typically, serum PTH increases near calving, activating bone resorption and renal tubular reabsorption to increase blood Ca (Goff, 2000). In the present study, this was not observed in the days examined (1, 3, 5, and 10 d postpartum). Kamiya et al. (2005) also observed no effect of dietary Ca on serum PTH. However, the researchers did report an effect of parity, with PTH concentration less in primiparous cows as compared with multiparous cows. This was not observed in the present study (8.3 ± 1.9 vs. 11.4 ± 2.3 pg/mL; primiparous and multiparous, respectively).
Bone Mineral Content
Bone data are presented in Table 6. Data are expressed as grams of Ca or P/sample in addition to percentage of wet weight, percentage of dry weight, and percentage of fat free ash weight. As bone resorption occurs through reduction of bone density (increased porosity) or bone wall thickness, loss in bone mineral mass will be reflected in a loss of mineral per unit of surface area. Thus, taking samples of fixed surface area and assessing for mineral mass in the total sample allows detection of loss of bone mineral from either reduction in density or thinning of the cross-sectional area.
|
). Dietary Ca concentration had no effect on bone Ca content on a mass or dry weight basis, but when expressed on a percentage of ash weight basis, the MED cows had lesser bone Ca content as compared with the LOW cows (Table 6). Bone Ca of cows consuming the HI diet was not different from the LOW or MED cows. The wk-11 bone sample for one cow on the LOW diet was identified as an outlier and removed from all bone analyses. Several covariates were tested to account for any possible difference in starting bone Ca concentration in the cows on the MED diet, including the 8 d sample, the preliminary OC and DPD concentrations, or the ratio of OC to DPD, but they were all nonsignificant. The SAS profile analysis was used to examine the difference between the samples. There were no significant differences for treatment with the profile analysis.
An unexpected result was the finding of greater bone Ca content for cows fed the LOW diet than for those fed the MED diet; nonetheless, the correlation of bone Ca and Ca balance was stronger for cows consuming the LOW diet (r = 0.56 vs. 0.37 MED and -0.06 HI). For the LOW cows, the strong correlation indicates that increased bone Ca was associated with a corresponding increase in Ca retention. However, for cows consuming the HI diet, Ca content in bone was not correlated to Ca retention. This is likely because they were consuming Ca in excess of their requirement. Excess Ca was excreted rather than used to build bone (Table 7).
|
Parity had no effect on bone Ca content but multiparous cows had greater bone P compared with primiparous cows on a percentage of wet and ash weight basis (Table 6). Wu et al. (2001) reported no differences in bone P as a percentage of ash (17.6% of ash) when cows were fed dietary P at 0.31, 0.39, or 0.47% for 2 or 3 yr. There was a linear increase in bone Ca content on a fat free ash basis across week (49.6, 52.8, and 54.8 ± 1.8% of fat free ash, for 8 d and 11 and 20 wk, respectively). However, there was a quadratic response in bone Ca grams/sample (0.23, 0.17, and 0.20 ± 0.007 for d 8 and wk 11 and 20, respectively). A quadratic response across time was also observed for grams/sample of fat free bone ash and P with the 8-d sample being greatest for both. The interactions of treatment x time and treatment x parity were not significant for bone Ca or P in grams/sample, or concentration on a wet, dry, or ash basis.
Ca Partitioning and Balance
There was a linear effect of dietary Ca treatment on fecal Ca excretion (Table 7); cows consuming more dietary Ca excreted more fecal Ca. Fecal Ca was different by week with linear and quadratic responses over time for all dietary treatments. There was a linear increase in fecal Ca excretion from wk 2 through the wk 11 balance. Week 11 had the greatest fecal Ca excretion across all treatments. This could be caused by a decrease in Ca binding protein (CaBP) potentially associated with estrus synchronization (discussed below; Inpanbutr et al., 1994). The decline in fecal Ca excretion at wk 20 caused the quadratic response. Multiparous cows excreted more fecal Ca as compared with primiparous cows (142 vs. 119 ± 4.3 g/d). This is likely the result of increased intake and is similar to the observations of Knowlton et al. (2001).
Urinary Ca excretion was not affected by diet (Table 7). Irrespective of treatment, all cows excreted the least urinary Ca during the wk-5 balance period, but urinary Ca was less than 1 g/d throughout. Primiparous cows excreted more Ca in their urine as compared with multiparous cows (0.90 vs. 0.57 g/d). Chan et al. (2006) reported no effect of prepartum dietary Ca (0.99 vs. 1.50%) on urinary Ca excretion and no effect of parity.
Milk Ca secretion was not influenced by dietary Ca treatment (Table 7). Multiparous cows had greater daily milk yield; therefore, multiparous cows also had greater daily milk Ca secretion as compared with primiparous cows (54.2 vs. 39.5 ± 2.2 g/d). The parity x time interaction was significant with multiparous cows decreasing in milk Ca secretion from wk 8 to 20, whereas primiparous cows did not change. This is the result of a significant decrease in milk yield in multiparous cows from wk 8 to 20; milk yield of primiparous cows remained constant in this time period. Weller et al. (2006) found that primiparous cows peaked in milk yield lower and later but had greater persistency of milk production as compared with multiparous cows. The interactions of treatment x week and treatment x parity were not significant.
As expected, Ca balance was affected by dietary Ca concentration (Table 7; Figure 4A). Cows consuming the LOW diet were in negative Ca balance for the entire 20 wk observed. During wk 2, cows consuming the MED diet had a negative Ca balance, but Ca balance was positive by wk 5. Knowlton and Herbein (2002) observed that all cows were in positive Ca balance by wk 5 with dietary Ca concentration at 0.72 to 0.77% of the diet DM. Cows consuming the HI diet had a positive Ca balance in most weeks except for wk 11, where all cows had a negative Ca balance due to an increase in fecal Ca excretion.
|
The positive Ca balance throughout the study in cows fed HI diets contradicts earlier work. In sheep fed a plentiful Ca diet, Ca balance was negative until 63 to 70 d of lactation (Braithwaite, 1983a). Others have concluded that cows are unable to absorb enough Ca to meet their demands in early lactation regardless of dietary Ca concentration (Braithwaite, 1983a; Horst et al., 1994). Our data at 2 wk postpartum does not support this. Cows consuming the HI diet with 1.03% dietary Ca apparently absorbed enough dietary Ca to meet their needs. However, our first balance period was not until 2 wk postcalving; Ca balance may well have been negative in all diets immediately postpartum.
The interaction of treatment x time was not significant for Ca balance (Table 7). The main effect of week was different with a positive linear and quadratic response. The quadratic response to dietary treatment may be explained by the dramatic drop at wk 11 followed by the increase at wk 20.
Phosphorus Partitioning and Balance
There was no effect of dietary Ca on fecal P excretion (47.5 ± 2.4 g/d), and the interactions of treatment x parity, treatment x week, and parity x week were not significant (Table 8). There was, however, both a linear and quadratic effect of week. Fecal P was least at 2 wk postpartum and greatest at wk 11 and 20 for all treatments. Apparent P digestibility was different over time with the greatest value of 37.3% occurring during wk 8 and 11 (quadratic response).
|
Most likely, however, the negative P balance was simply because these cows produced more milk than expected without concomitant increase in DMI. Diets were formulated for a cow consuming 23.2 kg of DM and producing 36.3 kg of milk per day, but the cows actually consumed 23.9 kg of DM and produced 41.0 kg of milk per day. The P requirement was calculated for these cows using their actual DMI and milk yield during each of the balance weeks and averaged 0.39, 0.41, and 0.42% of the diet for the LOW, MED, and HI, respectively. The difference in milk yield increased the requirement for absorbed P by 4.2 g/d.
Multiparous cows had greater fecal P excretion as compared with primiparous cows (53.6 vs. 41.1 ± 2.0 g/d) due to greater daily P intakes. These results are similar to Knowlton et al. (2001) where primiparous cows had lesser P intakes, reduced fecal P, and lesser milk P secretion. In contrast to the present study, Knowlton et al. (2001) observed an effect of parity on apparent P digestibility with primiparous cows having greater apparent P digestibility than multiparous cows.
Urinary P excretion was not affected by treatment or parity, and interactions of treatment x parity and treatment x week did not occur. The main effect of week was significant with a quadratic response for urinary P excretion (Table 8). The greatest urinary P occurred during the 2-wk balance (1.6 ± 0.3 g/d) followed by a decrease until wk 20 (1.2 ± 0.2 g/d). It is well accepted that P absorption, and in turn excretion, are directly related to P intake but that the kidneys are a minor route of P excretion (Hibbs and Conrad, 1983; Horst, 1986; Morse et al., 1992; Knowlton and Herbein, 2002). However, when bone mineral is resorbed in support of blood Ca concentrations, the urinary excretion of P can increase to maintain P homeostasis (Todd et al., 1962) if it is not needed. Wu et al. (2001) suggested that spilling of P into urine greater than 1 g/d per cow is a reliable sign that the animal has adequate P. The present study suggests otherwise because urinary P exceeded this threshold and all cows remained in a negative P balance (data below).
Dietary Ca concentration had no effect on milk P secretion (Table 8). Multiparous cows secreted more daily P in milk as compared with primiparous cows across all balance weeks with the exception of wk 20 where the 2 lactation groups were not different (parity x time). Milk P concentration, like milk Ca, is thought to remain relatively constant throughout lactation in ruminants (Gallego et al., 2006), so this is likely due to the decrease in milk yield in the multiparous cows during wk 20.
Phosphorus balance was not affected by dietary Ca concentration (Table 8). This does not support the original hypothesis. It was postulated that dietary Ca concentration would directly affect P balance in the lactating dairy cow when dietary P was held constant.
Whereas no effect of dietary Ca was observed, parity had a clear effect on these variables. Multiparous cows consumed more P, excreted more fecal P, and secreted more milk P, which resulted in a more negative P balance compared with primiparous cows (–13.69 vs. –7.98 ± 1.5 g/d).
Cows were in negative P balance regardless of dietary Ca concentration for the first 20 wk of lactation (Figure 4B). There was a quadratic effect of time with wk 11 approaching a 0 balance. This contrasts with the Ca balance data where the wk 11 decrease in Ca retention is the result of increased fecal Ca excretion. Knowlton and Herbein (2002) observed negative P balance until wk 7 when cows were fed a diet containing 0.34% P. However, over the entire 11-wk study the P retention was never above 5 g/d (Knowlton and Herbein, 2002).
Ekelund et al. (2006) attempted to force cows to utilize mobilized bone P with low dietary P (0.31%) in early lactation as compared with normal dietary P (0.43%). In that study, however, bone resorption and overall P retention were not different between the low and normal P dietary treatment groups; low dietary P did not induce increased bone resorption. Ultimately there was no change in P excretion between the 2 dietary groups. Similarly, Braithwaite (1983b) reported that in sheep, bone is mobilized in response to the animal’s needs for Ca and is not affected by dietary P content. The present study complements the study by Ekelund et al. (2006) in that varying dietary Ca concentration also did not affect P retention, bone resorption, or bone formation during the weeks examined.
Implications for P Feeding Recommendations
This study is unique in its examination of P requirements based on dietary Ca concentration to account for mobilized bone minerals. Bone that is resorbed is an available source of Ca and P that has not been entirely accounted for in current recommendations. There is not an environmental concern for Ca, but excess P in the diet can increase P in the manure. Therefore, if P requirements can closely match the animal’s needs and account for useable P coming from bone, the amount of P being excreted in manure and ultimately into the environment may be reduced. However, net times of bone resorption and formation need to be determined over the course of a lactation to ensure adequate mineral in the diet for when the animal switches from net resorption to formation of bone.
Regardless of treatment, data collected during balance weeks suggest that net bone formation occurred after the wk 5 balance based on ratios of OC to DPD across time. Irrespective of treatment, OC concentration was least during the wk-2 balance period and DPD concentration was greatest during wk 2 and 5. These relationships were inverted by the wk-8 balance. These changes may be indicative of a switch from net resorption to net formation of bone.
The changes in bone metabolism suggested by serum markers were not reflected in Ca and P balance. Calcium balance was greatest at wk 20 with values nearly double what was observed during all other balance weeks. Phosphorus balance was not different across time and was in fact negative during all weeks measured. Negative P balance cannot support net bone formation unless cows deposited bone mineral primarily as Ca, such as calcium carbonate. Our data support this hypothesis; Ca, but not P, content of bone samples increased over time and was greatest at wk 20 on a mass, wet weight, and ash weight basis. Also, bone Ca and P were not correlated, which further suggests that the bone mineral was something other than hydroxyapatite [Ca10(PO4)6(OH)2].
Bone biopsy data are useful in shedding light on changes in mineral metabolism, but the choice of sampling site influences conclusions. Benzie et al. (1955) examined which bones were the most sensitive to resorption in sheep consuming diets with differing Ca concentration and determined that bone resorption is not uniform within a bone and throughout the skeleton. The relative sensitivity of bones to resorption in descending order was vertebrae > pelvis > skull > sacrum > mandible > ribs > proximal end of tibia > scapula.
In the present study, the ribs were chosen for biopsy because of their accessibility and success of previous work utilizing such an approach to assess bone mineral status (Little, 1972; Beighle et al., 1993). However, it appears in the present study that Ca and P were being mobilized from bones other than the ribs, especially in the LOW Ca cows. From parturition until after wk 11, cows on the LOW diet were mobilizing between 10 and 30 g/d of Ca based on balance data. Through wk 20 they mobilized between 5 and 15 g/d of P. If cows were not severely deficient, the drafts on other bones were able to meet the demand for the minerals before the cow began resorbing the ribs. In contrast, Little (1972) reported when yearling cattle were fed a P deficient (P at 0.08% of diet) diet for 6 wk, rib bone had a greater bone P content at wk 1 than the 6-wk samples. Cattle were reported to be in negative P balance (1 to 2 g/d) over the entire study, but the methods for determining balance were not reported.
One of our objectives was to validate the use of OC and DPD as markers of bone metabolism with total collection and bone biopsies. Total collection is often used as the standard to determine mineral requirements but it is not error-free. There is an inherent compounding of error that occurs when calculating mineral balance from intake, feces, urine, and milk each with error that may lead to an increased likelihood of committing a type I error. Variability could potentially mask the physiological activity that is implied by the serum marker data. Alternatively, the markers of bone metabolism that were utilized are simply that: markers. Alone, they may not be a perfect method for determining Ca and P metabolism either. Based on our data, validation of these bone markers of metabolism with total collection and bone biopsy is conditional. During the wk-20 balance, there was greater Ca retention, bone formation (i.e., lesser DPD and greater OC), and greater bone Ca content as compared with earlier collections. This indicates that the markers of bone metabolism can be used as a noninvasive indicator of dietary effects on bone activity when the cow is in a state of metabolic stability (i.e., mid to late lactation).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication May 9, 2008. Accepted for publication July 18, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  V. R. Moreira, L. K. Zeringue, C. C. Williams, C. Leonardi, and M. E. McCormick Influence of calcium and phosphorus feeding on markers of bone metabolism in transition cows J Dairy Sci, October 1, 2009; 92(10): 5189 - 5198. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, or HI 1.03% 
) on serum Ca (closed symbols) and P (open symbols) concentration from –14 d prepartum to 140 d of lactation (A: SEM = 1.61; P < 0.72 for treatment x time interaction and B: SEM = 0.62; P < 0.27 for treatment x time interaction).
) cows. A: DPD = deoxypyridinoline, marker of bone resorption (SEM = 0.13; P < 0.006 for parity x time interaction) and B: OC = osteocalcin, marker of bone formation (SEM = 5.1; P < 0.23 for parity x time interaction).