HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Knowlton, K. F. Articles by Emmerson, D. A. Search for Related Content PubMed PubMed Citation Articles by Knowlton, K. F. Articles by Emmerson, D. A.

Animal management to reduce phosphorus losses to the environment¹

K. F. Knowlton^*,2, J. S. Radcliffe, C. L. Novak and D. A. Emmerson^,3

^* Department of Dairy Science, Virginia Polytechnic Institute and State University, Blacksburg 24060; and Department of Animal Sciences, Purdue University, West Lafayette, IN 47907; and Department of Animal and Poultry Science, Virginia Polytechnic Institute and State University, Blacksburg 24061; and and Aviagen NA, Huntsville, AL 35805

	Abstract

Top Abstract Introduction Role of Animal Agriculture Increasing Regulatory Pressure Phosphorus Requirements Reduced Overfeeding to Decrease... Feeding According to... Techniques to Improve... Improved Efficiency of... Economic Impact of Reducing... Implications Literature Cited

 
Water quality in the United States is threatened by contamination with nutrients, primarily nitrogen and phosphorus. Animal manure can be a valuable resource for farmers, providing nutrients, improving soil structure, and increasing vegetative cover to decrease erosion potential. At the same time, application of manure nutrients in excess of crop requirements can result in environmental contamination. Environmental concerns with P are primarily associated with pollution of surface water (streams, lakes, rivers). This pollution may be caused by runoff of P when application to land is in excess of crop requirements. Increased specialization and concentration of livestock and crop production has led to the net export of nutrients from major crop-producing areas of the country to areas with a high concentration of animal agriculture. Concentrated animal agriculture has been identified as a significant source of P contamination of surface water. Areas facing the dilemma of an economically important livestock industry concentrated in an environmentally sensitive area have few options. If agricultural practices continue as they have in the past, continued damage to water resources and a loss of fishing and recreational activity are inevitable. If agricultural productivity is decreased, however, the maintenance of a stable farm economy, a viable rural economy, and a reliable domestic food supply are seriously threatened. Decreasing the P content of manure through nutrition is a powerful, cost-effective approach to reducing P losses from livestock farms and will help farmers meet increasingly stringent environmental regulations. This paper reviews opportunities available to reduce the P content of livestock manure, including more accurate interpretation of the published P requirements of animals, improved diet formulation and group-feeding strategies to more precisely meet requirements, and approaches to improve availability of feed P for monogastric and ruminant species.

Key Words: Cattle • Phosphorus Excretion • Phytase • Poultry • Swine

	Introduction

Top Abstract Introduction Role of Animal Agriculture Increasing Regulatory Pressure Phosphorus Requirements Reduced Overfeeding to Decrease... Feeding According to... Techniques to Improve... Improved Efficiency of... Economic Impact of Reducing... Implications Literature Cited

 
Increased specialization and concentration of livestock and crop production has led to the net export of nutrients from major crop-producing areas of the country to areas with a high concentration of animal agriculture. Livestock utilize P inefficiently, excreting 60 to 80% of that consumed. Therefore, the majority of P brought on to the farm in feed stays on the farm, rather than being exported in meat or milk.

Animal manure is typically land-applied to supply nutrients for crop growth, but N and P are in imbalance in manure relative to crop needs. Land application of manure to meet the N needs of the crop results in the overapplication and accumulation of P in soils. Historically, P contamination of surface water was thought to be associated primarily with erosion. As application of P in excess of crop requirements continues, however, soil becomes saturated and runoff of P can occur independently of erosion (Daniel et al., 1992).

Areas facing the dilemma of an economically important livestock industry concentrated in an environmentally sensitive area have few options. If agricultural practices continue as they have in the past, continued damage to water resources and a loss of fishing and recreational activity are almost inevitable. If agricultural productivity is reduced, however, the maintenance of a stable farm economy, a viable rural economy, and a reliable domestic food supply are seriously threatened. Practices that reduce P losses from farms without impairing profitability must be developed and implemented.

Improved understanding of P digestion and metabolism in livestock will improve efficiency of P utilization, reducing P excretion and minimizing the imbalance of N and P in manure. This paper will review the requirements of dairy cattle, beef cattle, swine, and poultry species for dietary P, four nutritional approaches to reducing the P content of manure, and the economic impact of reducing P excretion and losses from the farm system.

	Role of Animal Agriculture

Top Abstract Introduction Role of Animal Agriculture Increasing Regulatory Pressure Phosphorus Requirements Reduced Overfeeding to Decrease... Feeding According to... Techniques to Improve... Improved Efficiency of... Economic Impact of Reducing... Implications Literature Cited

 
Concentrated animal agriculture has been identified as a significant source of P contamination of surface water (median contribution = 7 to 48% of total P loads, depending on watershed; Smith and Alexander, 2000). These losses were calculated with a model using measured stream water quality data and spatial data on sources, landscape characteristics, and stream properties. The relative importance of different nutrient sources varies greatly in different regions of the United States. Animal agriculture is a minor source of nutrient pollution in the populous Northeast and Great Lakes, for instance. In contrast, the Shenandoah Valley of Virginia is an example of an area of intensive animal agriculture associated with increased contamination of surface water. The Shenandoah Valley has the highest population of both dairy cattle and poultry in the state, and as many as 20% of the dairy farms also have at least one poultry house. Estimated manure nutrient production in the Shenandoah Valley exceeds crop requirements on a yearly basis. Manure phosphate per acre of cropland increased by 90% between 1978 and 1992, and an analysis of soil tests in 1993 and 1994 indicated that nearly 90% of samples were ranked "high" or "very high" in P (Pease et al., 1998). Although there is controversy as to the threshold level of soil test P that leads to P runoff, these soils clearly need no additional P.

The link between animal numbers, manure application to a limited land area, and P contamination of surface water was also demonstrated in the Lake Okeechobee watershed in Florida. From 1973 to 1988, P concentration in the water of Lake Okeechobee in Florida increased by 250% (Negahban et al., 1993). During this same time period, dairy cow numbers in the three counties surrounding the lake increased by more than 900 cows/yr (Boggess et al., 1997), and dairies were identified as the source of 40% of the P load to the lake (Negahban et al., 1993). The appearance of lake-wide algae blooms led to the imposition of stringent regulations designed to reduce agricultural runoff.

	Increasing Regulatory Pressure

Top Abstract Introduction Role of Animal Agriculture Increasing Regulatory Pressure Phosphorus Requirements Reduced Overfeeding to Decrease... Feeding According to... Techniques to Improve... Improved Efficiency of... Economic Impact of Reducing... Implications Literature Cited

 
Increasing public concern with water quality and increased awareness of the effects of concentrated livestock production have led to the development and implementation of increasingly stringent environmental regulations. Greater pressure on states from the federal government in the last decade to enforce federal clean water regulations has significantly increased the level of regulatory pressure felt by farmers.

One key change in water quality regulations in the past 5 yr is the shift from a primary focus on N to an increasing focus on P contamination of surface water. Limiting manure application to the P needs of the crop is one way to avoid continued accumulation of P in soil, and to minimize potential P runoff and contamination of surface water. Regulations limiting manure application to the P needs of the crop are in place for all farms in Maryland (Water Quality Improvement Act, 1998) and for poultry farms in Virginia (Virginia Poultry Waste Management Program, 1999). The recently finalized federal Concentrated Animal Feeding Operation regulations to address water pollution call for site-specific decisions on whether N- or P-based manure application limits are needed to protect water quality (EPA, 2001). Also, some federal cost-share funding is now being linked to the development and implementation of P-based nutrient management plans. Phosphorus-based nutrient management regulations dramatically increase the amount of land required to dispose of manure, and will have a severe, detrimental effect on the agricultural economy in areas of intensive animal agriculture.

	Phosphorus Requirements

Top Abstract Introduction Role of Animal Agriculture Increasing Regulatory Pressure Phosphorus Requirements Reduced Overfeeding to Decrease... Feeding According to... Techniques to Improve... Improved Efficiency of... Economic Impact of Reducing... Implications Literature Cited

 
Dairy Cattle
Phosphorus is required by lactating cows for bone mineralization or growth, milk secretion, energy metabolism, fatty acid transport, phospholipid synthesis, amino acid metabolism, and protein synthesis. Phosphorus is also a component of nucleic acids involved in cellular metabolism, enzyme systems, and buffer systems. Regulation of P balance involves absorption from the small intestine, mobilization from bone, and secretion in saliva. Phosphorus (phosphate) absorption in the small intestine increases on an absolute basis with increasing P intake despite a reduction in apparent digestibility of P in response to increasing dietary P content.

The National Research Council periodically reviews and summarizes the nutrient requirements of various species and issues publications listing these requirements. In the current dairy NRC, the P requirement is described using a factorial approach (NRC, 2001). The absorbed P requirement to support maintenance, growth, pregnancy, and lactation are calculated and summed, and then adjusted for availability of P in feedstuffs to calculate the amount of P that must be fed. Table 1 describes the total absorbed P requirement for nonpregnant, mature lactating Holstein cows of varying milk yield and DMI, and the dietary P concentration required for these nonpregnant mature cows. The P availability used to calculate this dietary requirement is from the sample lactating cow diet included in the software distributed with the publication.

Phosphorus Requirements for Beef Cattle
The P requirements for beef cattle in the latest NRC (1996) are calculated using the factorial method and are based on current body weight and rate of protein gain, milk production, and fetal weight in the last 3 mo of pregnancy for cows and heifers. The maintenance requirement is considered to be 16 mg of P per kilogram of body weight, reflecting endogenous fecal P loss. Phosphorus requirements for growth are calculated as 3.9 g of P per 100 g of protein gain, based on data published in 1950. Phosphorus needs for milk are similar to those in the dairy NRC, at 0.95 g/kg milk, and fetal P requirements are 7.6 g/kg fetal weight during the last 3 mo of gestation. Availability of feed P is assumed to be 68% from all sources. As an example, the dietary P requirements (grams/day) for growing and finishing Angus steers are presented in Table 2. The dietary P concentration needed to meet these requirements varies widely with DMI, breed, BW, growth rate, and physiological state.

Phosphorus Requirements for Swine
The P requirements for swine as reported in the latest swine NRC (1998) can be found in Table 3. Although the requirements listed in the NRC are the best available estimation of the requirement, it is important to realize that the accuracy of these requirements is limited by available data. In the latest swine NRC (1998), there were 151 references cited with regards to P requirements. Of those, only 37 were from the 1990s, and only 5 of those 37 addressed P requirements. Perhaps more importantly, only 1 of the 37 addressed the bioavailability of P from feed ingredients. Therefore, the majority of data used to support current estimations of P requirements is 15 yr old or older. As a result, the accuracy of these estimations may need to be questioned. Certainly the genetics and production characteristics of the U.S. pig population have changed drastically during that period of time. In addition, the genetics and production characteristics of grain sources used in swine diets have also changed.

Another consideration in formulating swine diets to meet P requirements is the available Ca (Ca_available) content of the diet. Ideally, diets should be formulated on an P_available:Ca_available basis, but data on the availability of Ca from feed sources for swine is almost nonexistent. The use of enzymes such as phytase alters the availability of P and Ca, further complicating the issue. These interactions will be discussed in more detail later in the paper. In the end, P nutrition for swine is an exercise in risk management, with the nutritionist trying to decide how low to reduce dietary P without affecting performance or carcass characteristics.

Phosphorus Requirements for Poultry
The most recent NRC publication addressing the nutrient requirements of poultry (NRC, 1994) has established a nonphytate P requirement of 0.45% for broilers from 0 to 21 d of age, declining to 0.30% during the finishing period. Higher requirements, particularly early in life, have been established for meat-type turkeys although controlled research regarding the requirements are even more limiting than for broilers. These recommendations are based on research published between 1952 and 1983 measuring growth rate, feed efficiency, and bone ash as response criteria. There has been very limited research regarding P requirements of broiler chickens in the last decade. Studies by Huyghebaert (1996, 1997), Beltran-Lopez et al. (2000) and Godoy et al. (2002) all confirm the NRC recommendation. However, a recent study also suggests that available P levels can be reduced by up to 30% with no negative impact on broiler production performance resulting in a significant reduction of P excretion (Fritts and Waldroup, 2003). A similar reduction in P excretion has been observed when feeding lower concentrations of dietary P to egg layers although the impact on egg production and quality has been inconsistent and compromised egg production and shell quality are sometimes observed (Summers, 1995; Keshavarz, 2000).

Skinner and Waldroup (1992) and Skinner et al. (1992) recently examined the influence of complete removal of inorganic P from broiler finisher diets on growth, bone strength, and the incidence of skeletal abnormalities. There were no negative effects of P removal on growth rate, feed conversion, tibia length or width, or the incidence of leg abnormalities in either of these studies. There was a reduction in bone breaking strength that could be largely eliminated through higher levels of Ca supplementation. More recently, Chen and Moran (1995) observed an increased incidence of skeletal defects following processing as a result of P withdrawal. However, the environmental potential of this P withdrawal justifies further research to identify more precise parameters that might allow significant reduction of P concentrations in broiler finisher diets. Although finisher diets are generally only fed in the last week before market, finisher feed typically represents approximately 40% of the total feed consumed by a commercial broiler.

The research cited above and used to establish the NRC requirements has focused on defining absolute requirements for nonphytate P. Many of these studies involved genetic stocks not common in the industry. This is of concern as animal genetics can influence requirements for P and responses to dietary additives such as phytase. Punna and Roland (2001) provided evidence that there may be variability in the response of different genetic strains to increasing available P in the diet. These researchers found a greater response to P and phytase supplementation in a low-susceptibility strain than in a high-susceptibility strain as indicated by the incidence of severe tibial dyschondroplasia lesions. This is an important observation because of the commercial and welfare importance of tibial dyschondroplasia. Several studies also suggest that the P requirement for maximal bone mineralization is higher than for growth (Huyghebaert 1996, 1997; Beltran-Lopez et al., 2000). These considerations become more important under industry conditions where nutrient availability is often limited due to feed quality, inconsistency of feed manufacture, management challenges, and compromised bird health including enteric health.

Studies regarding the impact of dietary phosphorus supplementation on bone mineralization and skeletal abnormalities also suggest that a holistic approach, which accounts not only for simple growth measures, but also health and metabolic end points and perhaps excretion should be adopted in establishing phosphorus requirements.

	Reduced Overfeeding to Decrease P Content of Excreta

Top Abstract Introduction Role of Animal Agriculture Increasing Regulatory Pressure Phosphorus Requirements Reduced Overfeeding to Decrease... Feeding According to... Techniques to Improve... Improved Efficiency of... Economic Impact of Reducing... Implications Literature Cited

 
In all species of livestock, P fed in excess of animal requirements is excreted, making reduced overfeeding a powerful tool to reduce the P content of manure.

Overfeeding and P Excretion—Dairy and Beef Cattle
In dairy cows, several studies indicate a direct link between P intake and P excretion (Morse et al., 1992b; Wu et al., 2000, 2001; Knowlton et al., 2001; Knowlton and Herbein, 2002). A Florida study was among the first to show this link (Morse et al., 1992b). Twelve cows were fed diets containing one of three concentrations of P (0.3, 0.41, 0.56% of dietary DM). Excretion increased linearly with increasing intake, and nearly all of the difference in P intake with the high-P diet compared to the low-P diet was excreted.

Overfeeding of dietary P is common in the field. Phosphorus is often fed to dairy cattle 20 to 40% in excess of published requirements (Shaver and Howard, 1995; Sink et al., 2000). A survey conducted by Wu (2003) in Pennsylvania indicates that the extent of overfeeding is less now than was indicated in these earlier surveys. In growing beef cattle, Erickson et al. (1999) reported an industry average dietary P content of 0.35 to 0.39% of DM, compared to published requirements for growing steers of about 0.20% P.

Why Overfeed?
In dairy cows, the most common explanation for overfeeding of dietary P is the perception that high-P diets improve reproductive performance. This perception likely originates from the observation that severe P deficiency impairs reproductive performance in cattle. The original studies that established this belief were primarily with range cattle (Eckles et al., 1932; Beeson et al., 1941), and the dietary P concentrations necessary to induce this impaired reproductive performance were below 0.20% of the dietary DM for lactating cows. This dietary concentration is far below the concentration found in most feedstuffs in modern lactating cow rations even without supplementation, and in all of these studies, P intake was seriously confounded with intake of energy and other minerals.

Although severe P deficiency may impair reproductive performance, there is no research data to suggest a benefit from feeding P to dairy cows in excess of NRC requirements (Brodison et al., 1989; Brintrup et al., 1993; Wu et al., 2000). A review by researchers in Wisconsin outlines 13 studies with 785 lactating cows fed diets low in P (0.32 to 0.40% P) or high in P (0.39 to 0.61% P; Satter and Wu, 1999). Dietary P had no effect on days to first estrus, days open, services per conception, days to first AI, or pregnancy rate (Table 4).

Field recommendations influence P intakes in the field. Inconsistent recommendations from nutritionists, veterinarians, and extension personnel have led many farmers to feed P in excess of the NRC recommendations. Until the environmental consequences became obvious, overfeeding P was viewed as cheap reproductive insurance. Revisiting the literature makes clear that there is no documented benefit to overfeeding P.

One final reason P is overfed is the inclusion of feeds in the diet that are naturally high in P. Many byproduct feeds are high in P, most notably the byproducts of corn processing and ethanol production. These are increasingly popular feed supplements for beef and dairy cattle because of the protein and energy they supply. However, inclusion of these feeds in higher amounts often increases the dietary P content beyond the animal’s requirement. Koelsch and Lesoing (1999) constructed nutrient balances for Nebraska livestock farms and found that producers who used these byproducts had greater imbalances between P inputs and outputs than producers who did not. In their study, the seven cattle operations who fed these products imported twice as much P onto their farms as they exported in meat, crops, and manure (input:output ratio of 2.0:1). In contrast, the nine farms that did not feed these products exported nearly as much P as they imported (input:output ratio of 1.1:1).

The popularity of these high-P byproducts will likely continue, and there is no easy solution to the problem of the resulting elevated dietary P. In the short term, producers using these feeds should at least remove unneeded supplemental inorganic P from diets. In the long run, however, the true cost of the use of these high-P feeds should be carefully considered. If the inclusion of these byproducts will cause significant nutrient imbalance in the livestock operation and lead to difficulty meeting environmental regulations, then these feeds may not be as inexpensive as they appear.

Overfeeding P and P Runoff.
Analysis of fecal samples from cows fed diets varying in P content indicates that increasing dietary P not only increases total fecal P, but also increases the amount and proportion of fecal P that is water soluble (Dou et al., 2002). Water-soluble P is the form of P that is most susceptible to loss in the environment. In cows fed diets containing 0.34, 0.52, or 0.67% P, the water-soluble fraction of fecal P was 2.9, 7.1, and 10.5 g/kg fecal dry matter, respectively. This increase in the proportion of water-soluble fecal P in cows fed high-P diets is likely because excess dietary P ends up in feces after being digested and absorbed from the small intestine. Absorbed P in excess of the animal’s requirements is re-secreted from the bloodstream into saliva (Ternouth, 1990) and then excreted in feces. By definition, this excess P is primarily in the water-soluble form.

Research in Wisconsin showed even more directly that addition of excess P to diets increases the potential for P losses from land-applied manure, even at similar manure P application rates (Ebeling et al., 2002). Manure from cows fed diets containing two different dietary P concentrations (0.31 vs. 0.49% P) were applied to silt loam soils and subjected to simulated rainfall just before corn planting and again after harvest. The manure from cows fed the high-P diet contained 2.7 times more P than the low-P manure. When applied at equivalent manure application rates, runoff of P from soils receiving high-P manure was 10 times higher than from soils receiving low-P manure. When manure application was adjusted for constant manure P application, P concentrations in runoff were four times higher from soils receiving high-P manure than from those receiving low-P manure despite equivalent P application rates.

Overfeeding and P Excretion—Swine
Swine manure contains high levels of P, the result of poor digestibility of phytate P and feeding above the P requirement. In pigs as in cows, overfeeding P increases P excretion directly. Bridges et al. (1995) fed 24 crossbred barrows diets containing 0.50, 0.40, or 0.30% P plus phytase at 1,000 units/kg. Twelve pigs were lightweight, averaging 61.8 kg at the start of the study, and 12 were of heavyweight, averaging 101.2 kg at the start. The high-P diet was typical of that fed to finishing pigs at the time of the study, whereas the 0.40% P diet reflected requirements according to the 1988 NRC. In the lightweight pigs fed the low P–phytase diet, the intake of P was reduced by 4.08 g/d per pig (–37.3%) compared to high-P control, and P excretion was reduced by 2.94 g/d (–40%). In the heavyweight pigs, the reduction in P intake per pig was 4.53 g/d (–34%) and the reduction in P excretion was 3.66 g/d (–37.8%) with the low P–phytase diet compared to the high-P diet. In both studies, the apparent digestibility of P and P retention were not affected by dietary P content (P > 0.10).

Kornegay and Verstegen (2001) presented data from several surveys conducted in the late 1980s through the mid-1990s, indicating that P was being fed at 110 to 155% of its requirement as listed in the most recent NRC (1998). The result of this overfeeding is a diet that is more expensive and results in a larger concentration of P in manure. As in dairy cattle diets, overformulation of P in swine diets results from uncertainty of the actual P requirement, lack of homogeneity of the group of pigs being fed, and misinterpretation of published requirements.

Advances in research are improving this situation, in swine as in other species. Jongbloed et al. (1997) estimated that P excretion per growing pig in the Netherlands was reduced by 50% between 1973 and 1995 because of improvements in the understanding of P availability and P requirements and the implementation of exogenous phytase in pig diets. This reduction in excretion was due to the reduction in dietary P content of growing-finishing pigs in the Netherlands by more than 2.5 g/kg and a simultaneous increase in the efficiency of feed conversion.

Overfeeding and P Excretion—Poultry
As in swine and cattle, intake and excretion of P are tightly linked in broilers, layers, and turkeys. Summers (1995) fed White Leghorn hens diets varying in P_available content from 18 to 64 wk of age. The low-P diet (0.2% P_available) was below NRC requirements, the medium-P diet (0.3% P_available) was similar to required, and the 0.4% P_available diet was considered a "conventional" control diet. Phosphorus excretion was reduced by 20 and 40% in birds fed diets containing 0.3% P_available and 0.2% P_available, respectively, compared to birds fed the control diet. Egg production was reduced in birds fed the low-P diet, but egg production, egg weight, and shell quality were similar in birds fed 0.3% and 0.4% P_available. The reduction in P excretion with the 0.3% P_available diet was identical to the reduction in the total P content of the diet.

The published dietary P requirements appear to be recognized as acceptable based on a lack of recent publications on the subject and the fact that industry feeding practice is broadly consistent with these standards (Figure 1). Thus, efforts to improve P availability and utilize more highly available P sources would appear to have greater potential to reduce P excretion than focusing on overfeeding itself.

View larger version (11K): [in this window] [in a new window]   Figure 1. Comparison of current industry practice with NRC (1994) recommendation for available P (Pavailable) in broiler diets.

	Feeding According to Physiological State

Top Abstract Introduction Role of Animal Agriculture Increasing Regulatory Pressure Phosphorus Requirements Reduced Overfeeding to Decrease... Feeding According to... Techniques to Improve... Improved Efficiency of... Economic Impact of Reducing... Implications Literature Cited

 
In all species, nutrient requirements vary with physiological state, including changes in growth rate, productivity, pregnancy, gender, and age. Formulating diets to meet the P requirements of a more homogenous group of animals reduces overfeeding, reducing the P content of manure.

Crediting Bone P Resorption in Early Lactation Cows
In lactating ruminants, opportunity exists to reduce P excretion by accounting for the P released through the normal catabolism of bone that occurs in early lactation. When sheep were fed diets varying in Ca and P content throughout pregnancy and lactation (Braithwaite, 1983b), body P reserves were mobilized in late pregnancy and early lactation. This mobilization was in response to Ca requirements and was not affected by dietary P content.

Bone consists of Ca and P deposited within an organic collagen matrix. The highly porous nature of this matrix provides bone with an extensive surface area, making bone a highly labile source of both Ca and P. Bone mineral content is the result of the balance between rates of bone accretion and bone resorption. Bone accretion remains relatively constant within animals of a given age (Braithwaite, 1976; Braithwaite, 1983a), so changes in bone mineral retention are due primarily to changes in rate of resorption.

Net bone resorption is likely a normal consequence in early lactation due to the rapid increase in demand for Ca to support milk yield. The postpartum ruminant is typically in negative Ca balance (Braithwaite, 1976), and parathyroid hormone is secreted in response to this hypocalcemia (Shappell et al., 1987). Parathyroid hormone stimulates the conversion of 25-hydroxycholecalciferol to dihydrocholecalciferol. Together, parathyroid hormone and dihydrocholecalciferol stimulate bone resorption (Braithwaite, 1976; Ternouth, 1990).

Phosphorus release from bone during early lactation provides a readily available source of P to meet the needs for maintenance and milk yield. Ternouth (1990) suggested that beef steers fed P-deficient diets could mobilize up to 30% of bone mineral, or 6 g/d of P, meeting about half of their dietary requirement. Satter (2002) extended this estimate to a 600-kg lactating cow, and estimated that as much as 600 to 1,000 g of P could be mobilized in early lactation. Supporting this, Knowlton and Herbein (2002) observed that mobilization of P from body reserves may meet a significant proportion of the dairy cow’s net need for P during early lactation. Assuming P balance reflects P resorption from bone, cows mobilized up to 25 g/d of P from bone in the first 3 to 5 wk of lactation. The requirement for absorbed P in early lactation totals 45 to 70 g/d, depending on milk yield (NRC, 2001). Although Knowlton and Herbein (2002) reported no effect of diet on P retention, they observed an interaction between week of lactation and dietary P content, suggesting that dietary P affected the duration of net bone P resorption. Cows in all treatment groups mobilized P from body reserves at wk 3 of lactation, but only cows fed the low-P diet remained in negative P balance during wk 5. Reserves of P mobilized in early lactation must obviously be replenished in later lactation, and questions remain about the timing and rate of replenishment of bone P stores, as well as the duration and ultimate extent of mobilization of those reserves.

Phase Feeding—Swine
The P requirement of the pig changes with age. With increasing age, the quantity of P required increases, but, due to increased feed intake, the required P content of the diet decreases (NRC, 1998). The more frequently the diet is reformulated, the more accurately the nutrient needs of the animal can be met, and increased efficiency of nutrient utilization will result in decreased excretion.

Phase feeding is a management tool for reducing nutrient excretion, and it is commonly practiced in the industry today. However, very little if any recent research has focused on the reduction in P excretion as a result of phase feeding. Several studies have documented substantial reductions in N excretion as a result of phase feeding (Henry and Dourmad, 1993; Chauvel and Granier, 1996; Vander Peet-Schwering et al., 1996). Using estimates of P requirements during the growing and finishing periods (NRC, 1998) and estimates of feed efficiency (Dritz et al., 1997), the reduction in P consumption per pig as a result of phase feeding can be calculated. The P consumption over the entire grow-finish period can be estimated by summing the P intakes for each phase, which can be estimated by dividing the BW gain for each phase by the gain:feed ratio and multiplying by the P content of the diet ([BW gain/gain:feed] x dietary P). Assuming one dietary phase was used, and P was held constant at 0.50% of the diet to meet the requirements of the youngest group, P intake would be estimated as

If, instead, a three-phase feeding system was implemented, the reduction in P intake and excretion can be calculated. Assuming total P requirements of 0.50, 0.45, and 0.40% and gain:feed estimates of 0.424, 0.341, and 0.274 for pigs weighing 20 to 50, 50 to 80, and 80 to 120 kg, respectively, P consumption can be estimated as

This represents a 12.5% decrease in P consumption and a corresponding reduction in P excretion.

Similar to phase feeding, diets can be formulated to more precisely meet the needs of animals for P if animals are grouped and fed by gender, as well as by age. Gilts consume less feed and are leaner at the same BW compared to barrows (Cromwell et al., 1993). Therefore, the dietary concentration of P and other nutrients can be reduced in diets for barrows compared to equal-BW gilts. An overall improvement in feed efficiency will be observed because barrows are not being overfed in order to meet the nutrient requirements of gilts. Calculations similar to those made above for phase feeding can be made for split-sex feeding. Ultimately, phase feeding and split-sex feeding will result in decreased total consumption of P, increased efficiency of dietary P utilization, and decreased P excretion.

Phase Feeding—Poultry
Recent research regarding phase feeding for poultry has primarily focused on amino acid nutrition. However, Huyghegaert (1996 However, Huyghegaert (1997) reported comprehensive research regarding phase feeding for P nutrition. These studies provided response surfaces for Ca, P, and phytase supplementation for both growth and bone mineralization and then demonstrated that higher Ca-to-P ratios were beneficial for maximizing bone mineralization and reducing the incidence of tibial dyschondroplasia. The author also suggested that dietary Ca and P could be reduced significantly by feeding a two-feed program to broilers. This concept is reflected in both the NRC recommendation and current industry feeding practice (Figure 1). The situation is even more extreme in turkey production, where dietary P concentration is reduced by more than 250% over the 20-wk growing period. Current poultry industry feeding practice includes the phase feeding concept where broilers are fed as many as five diets to 40 d of age and turkeys consume up to nine different diets to 20 wk of age.

	Techniques to Improve Availability of Feed Phosphorus

Top Abstract Introduction Role of Animal Agriculture Increasing Regulatory Pressure Phosphorus Requirements Reduced Overfeeding to Decrease... Feeding According to... Techniques to Improve... Improved Efficiency of... Economic Impact of Reducing... Implications Literature Cited

 
The availability of P in feedstuffs affects total P requirements in all species, but our assumptions of the availability of feed P are based on relatively few studies. Questions remain in all species about appropriate methods and response variables to determine the availability of feed P. Improved P availability from feed allows the tissue-level needs of the animal to be met with reduced P intake, reducing the P content of excreta.

Approximately 65 to 70% of the total P in cereal grains is organically bound in phytate P (Nelson et al., 1968; Morse et al., 1992b). This form of P is relatively unavailable to monogastric animals as they lack the enzyme phytase (Cromwell, 1992; Ravindran et al., 1994, 1995). Phytase catalyzes the release of phosphate groups from the inositol ring of phytate, making the P more available for absorption in the small intestine (Jongbloed et al., 1992; Cromwell et al., 1993; Lei et al., 1993b; Kornegay, 1995, 1996; Jongbloed et al., 1996; Radcliffe and Kornegay, 1998; Skaggs, 1999; Rice et al., 1999; Zhang, 1999; Rice et al., 2000; Robbins et al., 2000). Phytate P is more available to ruminants than to nonruminants because ruminal microorganisms possess phytase (Clark et al., 1986; Morse et al., 1992b; Yanke et al., 1998).

Improved Availability of Feed Phosphorus in Swine Diets
In the swine industry, there is a great deal of interest in lowering the dietary concentration of phytate P, making more of the plant P available to the pig for absorption. There have been two primary strategies employed to accomplish this: the addition of exogenous phytase to the diet and the development of plants with reduced phytate P content. Both methods result in an increased efficiency of P utilization and decreased P excretion. This section will focus on the use of exogenous phytase in swine diets because this method is commercially available at the present time. However, low-phytic acid grains do have the potential for wide-scale use in the future if agronomic issues such as reduced crop yields can be overcome. In addition, the combination of low-phytic acid grains and phytase may ultimately provide the producer with the greatest ability to lower dietary P content.

Site of Phytase Activity.
To assess the site of activity of exogenous phytase within the digestive tract, Jongbloed et al. (1992) studied pigs fitted with simple T-cannulas in the duodenum (approximately 25 cm posterior to the pylorus) and ileum (approximately 15 cm anterior to the ileocecal junction). The pigs were fed a diet containing exogenous phytase, and 85% of the ingested phytase activity was recovered in duodenal digesta, whereas no phytase activity was recovered in ileal digesta. They concluded that the primary site of action of exogenous phytase was in the stomach and duodenum. Confirming this, in two experiments, Yi and Kornegay (1996) reported that phytase activity in the digesta, measured 3 h after ingestion of a meal, decreased from the stomach to the upper small intestine to the lower small intestine. Phytase activity, as a percentage of the total consumed phytase activity, was found to be 51% in the stomach, 31% in the upper small intestine, and 5% in the lower small intestine in Exp. 1. In Exp. 2, values of 41, 16, and 5% were reported for the stomach, upper small intestine, and lower small intestine, respectively.

Site of activity of exogenous phytase is influenced by the digesta pH. The pH optimum for commercially available phytases varies from 2.5 to 5.5; at pH greater than 6, activity is essentially lost. The acidity of the stomach lumen ranges from pH 1.0 to 4.5 (Chesson, 1987) and the luminal pH of the gastrointestinal tract increases from the duodenum to the terminal ileum. Duodenal pH in the pig averages approximately 4.8 but varies with time postfeeding. Jongbloed et al. (1992) reported that the duodenal pH immediately following a meal was 5.7, after which it gradually decreased to pH 3.3. The jejunum, which represents approximately 90% of the total length of the small intestine, has a mean pH of 5.5 to 6.9, whereas the ileum has a mean pH of 7.0 to 7.4.

Effects of Exogenous Phytase on Phosphorus Digestibility.
Nelson et al. (1968) demonstrated the ability of exogenous phytase to release P bound to phytic acid. However, only recently has a commercially available phytase preparation been approved for use in swine diets. Addition of exogenous (microbial) phytase has been shown to catalyze the hydrolysis of the phytate molecule, releasing bound phytate P (Jongbloed et al., 1992; Cromwell et al., 1993; Lei et al., 1993b, Kornegay, 1995, 1996; Jongbloed et al., 1996; Radcliffe and Kornegay, 1998; Skaggs, 1999; Rice et al., 1999; Zhang, 1999; Rice et al., 2000; Robbins et al., 2000).

Microbial phytase has been shown to affect performance of pigs fed low-P diets by increasing average daily gains primarily due to an increased feed intake (Simons et al., 1990; Jongbloed et al., 1992; Kornegay and Qian, 1996; Yi et al., 1996). Increases in bone breaking strength or shear force have also been observed in several studies with pigs (Cromwell et al., 1993; Kornegay and Qian, 1996; Yi et al., 1996; Radcliffe and Kornegay, 1998) when phytase was added to a low-P diet. The addition of microbial phytase also decreases P excretion by 25 to 50% (Simons et al., 1990; Jongbloed et al., 1992; Cromwell et al., 1993a; Lei et al, 1993b; Kornegay and Qian, 1996; Yi et al., 1996) by increasing P digestibility or retention (Hoppe et al., 1992; Lei et al, 1993a, b; Mroz et al., 1994; Kornegay and Qian, 1996; Yi et al., 1996) and by decreasing the need for supplemental inorganic P.

Phosphorus Equivalency Values of Phytase.
For swine producers to efficiently use microbial phytase, accurate equivalency values of phytase for P must be developed. Ideally, in studies designed to develop equivalency values, diets varying in dietary P content should be fed without added phytase and multiple levels of phytase should be fed with a low-P diet to develop response equations for both P and phytase. These response equations can then be set equal to one another and solved to determine the P equivalency of phytase. In general, linear (Y = a + bX, where Y = response and X = the level of P or phytase) or asymptotic (Y = a(1 – be^–kX), where Y = response and X = the level of P or phytase) curves have provided the best fits for phytase and P responses in corn/soybean meal-based diets (Kornegay et al., 1998). Jongbloed et al. (1996) reported that a logistic curve provided a better fit to the response of P and phytase in a practical Dutch diet.

Before examining the results of several studies attempting to generate P equivalency values, it is first necessary to distinguish between the amount of phytate P released by phytase and P equivalency values of phytase for inorganic P. Many studies in the literature do not attempt to distinguish between these, but they are quite different. The amount of phytate P released by phytase refers to the amount of P released due to the hydrolysis of phytate P. This value is lower than the P equivalency value because phytase is replacing inorganic P, which is less than 100% available. Therefore, in order to determine the amount of inorganic P that can be replaced in a given diet, it is essential to know the bioavailability of P from the inorganic P source. For example, if 500 U of phytase per kilogram of diet could catalyze the release of 0.76 g of phytate P, then the P equivalency value would be equal to 0.76 g divided by the percentage availability of the inorganic P source. If the P from inorganic P was 75% available, then the P equivalency value of phytase would be 1.01 g of P from inorganic P (0.76/0.75).

The bioavailability of P from inorganic P sources used in swine diets is generally quite high, but there has been a substantial amount of variation reported in the literature. Soares (1995) suggested a relative bioavailability value of 90% for defluorinated phosphate, 95% for dicalcium phosphate, and 100% for monocalcium phosphate when monosodium phosphate was used as the standard and given a value of 100. A relative bioavailability of 87% is suggested for defluorinated phosphate and a value of 100% is suggested for dicalcium phosphate when monocalcium phosphate is given a value of 100 (NRC, 1998). Kornegay and Radcliffe (1997) compared four sources of defluorinated phosphate and one source of dicalcium phosphate against a monocalcium phosphate control. They found no significant differences among any of the phosphate sources, with relative bioavailabilities ranging from 95.1 to 105.3%. These relative bioavailabilities were converted to availability of each P source in this study by dividing the increase in digested P (g/kg; digested P in diet – basal digested P) when P is added by the amount of added inorganic P added and multiplying by 100. With the exception of one of the defluorinated phosphate sources (bioavailability = 58.5%), the estimated bioavailabilities of P from all sources used in the study by Kornegay and Radcliffe (1997) ranged from 71.9 to 79.3%, with an average of 75.1%. This is in good agreement with a review by Kornegay et al. (1998) in which the estimated bioavailability of inorganic P across 52 experiments was 76.7% for swine.

For pigs, the P equivalency value for 500 U of phytase per kilogram of diet reported in the literature ranges from 0.27 to 2.47 g of P (Tables 8 and 9). Factors that may influence these equivalency value estimates include grain types fed, the P and phytate content of the diet, response criteria used and, perhaps most importantly, the ratio of Ca to P. Phosphorus absorption is impaired in pigs if the Ca-to-P ratio is too wide (NRC, 1998). In addition, Qian et al. (1996) reported a detrimental effect of a widening Ca-to-P ratio in excess of 1.2:1 on phytase efficacy in pigs. This observation may be the result of a poorer P absorption due to a wide Ca-to-P ratio rather than a decrease in enzyme effectiveness. Excess Ca may also bind to the phytate molecule, making it insoluble and therefore unavailable for exposure to phytase in the gastrointestinal tract.

Radcliffe and Kornegay (1998) fed 96 crossbred weanling pigs a low-P diet (3.5 g/kg) with supplemental inorganic P (0, 0.5, 1.0, or 1.5 g/kg) or microbial phytase (0, 167, 333, or 500) to determine the P equivalency value of microbial phytase. They found that the addition of microbial phytase to a low-P diet improved ADG, rib shear force, shear energy and ash (% and weight), and Ca and P digestibility. The addition of P to the low-P diet improved ADG, rib shear force, energy and ash (% and weight), and Ca and P digestibility. After regressing response criteria against P or phytase level, linear (Y = mX + b) P response curves provided good fits for P digestibility and rib shear force. Nonlinear (Y = a(1 – be^–kX)) P response curves provided good fits for ADG and rib ash weight. Phytase addition resulted in linear increases in ADG, rib ash weight, and shear force. Nonlinear increases in P digestibility were observed. By setting the P response equations equal to the phytase response equations and solving for P, phytase equivalency values were derived. In this study, 500 U/kg of phytase was equivalent to 0.84 g of P as inorganic P.

Kornegay et al. (1998) used data from 52 pig experiments to estimate the P equivalency value of phytase. Data from these experiments were used to generate response curves for P digestibility (%), digested P (g/kg), and P excretion as influenced by phytase and P. Figure 2 shows the phytase response curve that was developed for P digestibility. Phytase unquestionably increases P digestibility. However, the magnitude of this response is dependent on diet type, total P content of the diet, phytate P content of the diet, Ca-to-P ratio, and the age and physiological status of the pig. These criteria account for the wide range of P digestibilities reported in the literature. The P digestibility curve described for phytase by Kornegay et al. (1998) was plotted with the actual data used to generate the curve on the same graph in Figure 2. Examining the basal diets, with no added phytase, a range of P digestibility from 8.4 to 63.0% was observed. This wide range is due to the inclusion of plant feedstuffs with intrinsic phytase activity in some studies, differences in the Ca-to-P ratio, differences in the inclusion level of inorganic P, and differences in the phytate P level of the diet. This variation continues as phytase is added to the diet and causes the relatively poor fit (r² = 0.47) of the response curve calculated in this review.

View larger version (20K): [in this window] [in a new window]   Figure 2. Apparent P digestibility response curve adapted from Kornegay et al. (1998). (Data used to derive the response curve were obtained from Eeckhout and de Paepe, 1991; Nasi, 1991; Beers et al., 1992; Jongbloed et al., 1992, 1996; Lei et al., 1993a,b; Mroz et al., 1993, 1994; Adeola and Sutton, 1995; Cromwell et al., 1995; Kornegay and Qian, 1996; Han et al., 1997; Harper et al., 1997; Liu et al., 1997; Murry et al., 1997; and Kornegay, et al., 1998.)

View larger version (21K): [in this window] [in a new window]   Figure 3. Increase in apparent P digestibility (percentage unit) response curve calculated from Kornegay et al. (1998). (Data used to derive the response curve were obtained from Eeckhout and de Paepe, 1991; Nasi, 1991; Beers et al., 1992; Jongbloed et al., 1992, 1996; Lei et al., 1993a,b; Mroz et al., 1993, 1994; Adeola and Sutton, 1995; Cromwell et al., 1995; Kornegay and Qian, 1996; Han et al., 1997; Harper et al., 1997; Liu et al., 1997; Murry et al., 1997; and Kornegay et al., 1998.)

View larger version (18K): [in this window] [in a new window]   Figure 4. Comparison of phytase response curves for the increase in apparent P digestibility (%). The response curve of Kornegay et al. (1998) was based on the P digestibility response curve presented in that paper. (The response curve of Dungelhoef and Rodehutscord was based on data from Nasi, 1990; Simons et al., 1990; Borggreve et al., 1991; Beers, 1992; Beers et al., 1992; Eeckhout and De Paepe, 1992a,b,c; Jongbloed et al., 1992; Lantzsch and Wjst, 1992; Jongbloed et al., 1993; Kemme and Jongbloed, 1993; Mroz et al., 1993, 1994.) The response curve of Jongbloed (1996) was based on data from Beers et al., 1992.

View larger version (18K): [in this window] [in a new window]   Figure 5. Comparison of phytase response curves for the increase in apparent P digested (g/kg of diet). The response curve of Kornegay et al. (1998) was based on the P digestibility response curve presented in that paper. (The response curve of Dungelhoef and Rodehutscord was based on data from Nasi, 1990; Simons et al., 1990; Borggreve et al., 1991; Beers, 1992; Beers et al., 1992; Eeckhout and De Paepe, 1992a,b,c; Jongbloed et al., 1992; Lantzsch and Wjst, 1992; Jongbloed et al., 1993; Kemme and Jongbloed, 1993; Mroz et al., 1993, 1994.) The response curve of Jongbloed (1996) was based on data from Beers et al. (1992).

View this table: [in this window] [in a new window]   Table 10. Equations used to generate phytase response curves in Figures 3 through 5 and response values for 250, 500, and 1,000 U of added phytase activity per kilogram of diet as predicted by the equations used to develop the curves

Kornegay (1999), in a review of literature, estimated that the average bioavailability of P from multiple inorganic P sources was 46.2% for broilers. For comparative purposes, the relative bioavailability of P in calcium phosphate supplements is generally set at 100%, with the bioavailability of other sources expressed relative to this. The P from rock phosphate sources, for instance, is less available than the P from calcium phosphate (Sullivan et al., 1994; Summers, 1997). The bioavailability of P from commercial dicalcium phosphate products varies considerably, however, and depends on the specific response criteria evaluated (Lima et al., 1997). Phosphorus bioavailability was greater and less variable when evaluated based on growth effects and relative blood P concentrations as compared to bone ash and bone breaking strength criteria. Relative P bioavailability ranged from 79 to 110% for the seven commercial sources of dicalcium phosphate examined (including several of non-U.S. origin). Thus, there is considerable variation in the availability of P for even the most highly available inorganic source