Effect of Strain and Finisher Diet Non-Phytate Phosphorus Level on Performance and Litter Composition in Large Tom Production

There are several challenges associated with maintaining profitability in large tom turkey production. Among these are choosing the commercial genotype or strain that optimizes feed conversion ratio (FCR), ending weight (EW) and breast yield according to each company’s market needs. There have been concerns with “skeletal weakness” in commercial toms for over 40 years. In 1971 the National Turkey Federation sponsored a symposium at Iowa State University to address this and the same correlated factors we recognize today were discussed then (i.e. litter condition, skeletal pressures due to rapid growth, diet). The cost of turkey research combined with a significant decline in the number of poultry research institutions have contributed to commercial diets often containing levels of total P and nPP that are well in excess of NRC (1994) recommendations. This comes at a time when there is also increased attention on the environmental consequences of using poultry litter as fertilizer. Much of the concern is due to the potential impact of manure phosphorus (P) on the environment. The contribution of P from agricultural soils to the potential for eutrophication within large bodies of water has changed our dietary management practices with respect to mineral levels. Due to the perception that land application of manure is a primary point source of P, strategies to minimize manure P have become a priority for poultry producers.

Dietary P is essential for growth and adequate skeletal development and this is particularly important for rapidly growing heavy toms (Waldroup, 1999). The NRC (1994) recommendations for available P or what we may best estimate analytically, non-phytate phosphorus (nPP), have not changed since 1984 although the market weight of toms has increased from approximately 14 kg to 19 kg at 19 wk of age. Roberson et al. (2004) reported that toms fed diets containing 145% versus 100 or 75% of the NRC (1994) recommended levels of Ca and nPP from 9 to 17 week had significant increases in tibia strength and bone ash at 15 and 17 week of age but also increased litter P. The inclusion of nPP containing ingredients in the diet is expensive but necessary due in part to the fact that the digestibility of P in common feed ingredients is highly variable (Van der Klis and Versteegh, 1996). The chemical definition of nPP is the difference between the analyzed phytate phosphorus (PP) content of an ingredient or feed and the analyzed total P content (Angel and Applegate, 2001). It has been determined that 28.2% of the phytate molecule is P (Angel et al., 2002) and nPP is more readily digested compared with bound PP. While it is well accepted that reducing dietary P will contribute to reduced litter P, the accelerated growth of commercial toms and the importance of P to skeletal integrity can influence both the welfare of a flock in addition to mortality and feed conversion during the latter stages of growth (Rath et al., 2000). The cost of leg problems combined with the dearth of published data on phosphorus requirements of commercial toms have contributed to fairly high safety margins for dietary P in commercial diets, particularly for older toms. Phase feeding has been implemented in commercial turkey production to meet a bird’s nutrient requirements with specific diet formulations based on body weight and age. The finisher phase represents an opportunity to reduce dietary nPP and litter P because skeletal growth has largely stopped (Turner and Lilburn, 1992) and the finisher period represents approximately 50-60% of total feed consumed. The objective of this research was to determine strain and finisher diet nPP level effects on performance and litter composition of large tom turkeys.

Male poults from two commercial strains (A = Nicholas Turkeys, Aviagen Group, EW Group, Visbek, Germany; B = Hybrid Turkeys, Hendrix Genetics, Boxmeer, the Netherlands) were used in each of two experiments. All poults were randomly allocated at d 1 to 16 pens (6.1×5.2 m) in the West Virginia University, Reymann Memorial Farm, turkey research facility. All pens contained fresh wood shavings on top of concrete flooring. The barn utilized tunnel ventilation and radiant brooders. Feed and water were provided for ad libitum consumption with bell drinkers and an augered feed pan system. In addition, each pen contained an automated bird and feed dump scale (Chore-Time Brock Incorporated, Milford, IN). Both Exp contained 8 replicate pens specific to strain; however, the number of poults initially placed per pen was 76 and 90 in Exp 1 and 2, respectively. The temperature and lighting programs followed standard industry protocols (Virginia Poultry Growers Cooperative, 2009-2011).

Within each experiment, a typical industry 6 phase feeding approach was used where 8 diets were fed. For Exp 1, the diet of experimental interest (Finisher 2) was fed from 113 to 136 d while in Experiment 2, Finisher 1 was fed from 78 to 112 d and Finisher 2 from 113 to 126 d. Pelleted diets were fed during all phases except starter 1 and 2 that utilized crumbled feed. Diets were formulated to industry standards and dietary treatments were only applied to the finisher phase diets which contained either normal (Norm) or Low dietary nPP levels. All diets were manufactured at a commercial feed mill (Virginia Poultry Growers Cooperative, Broadway, VA). Diets contained corn, soybean meal, poultry by-product meal, wheat middlings, animal/vegetable blended fat, tricalcium P, monocalcium P and a commercial phytase. Diet formulations were proprietary; however, P assays were conducted on feed samples (NP Analytical Laboratories, St. Louis, MO). In addition, proximate analysis data was obtained for diets fed in Exp 2 (NP Analytical Laboratories, St. Louis, MO). The low nPP diets for both Exp 1 and 2 were created by decreasing and/or removing supplemental inorganic P (i.e. tricalcium P and monocalcium P). Phytase (1000 FTU/kg) was included in both finisher diets and our assumption was a release of 0.1% nPP. In Exp 1, the analyzed total P levels in the finisher 2 diets were 0.55% (Norm) and 0.46% (Low) with corresponding calculated nPP levels of 0.37% (Norm) and 0.31% (Low), respectively. Thus, the calculated nPP excluding phytase release was reduced 19% for the 23 d experimental period. The Exp 2 diet analysis and P assay results are shown in Table 1. The nPP levels were reduced by approximately 5% in the norm and low diets over the entire 48 d experimental period, again excluding phytase release. The lack of similarity of nPP reduction between studies may be associated with a compounding calculation error of total P and phytic acid analysis or the fact that these diets were prepared in a commercial feed mill during a standard work day, that would likely decrease batching/mixing precision relative to a research oriented feed mill.

Performance data including EW, FCR and pen mortality percentage were measured weekly throughout both studies. Automated bird and feed scales were calibrated routinely and individual bird weights were manually taken twice during the study (d 42 and d 136 and 126 in Exp 1 and 2, respectively) to confirm the automated bird scale accuracy. Litter samples were collected from standardized locations within each pen and then pooled prior to analysis. The pooled sample from each pen was analyzed for total P and inorganic P (New Jersey Feed Laboratory Incorporated, Ewing, NJ). Inorganic P was determined because this form of P is most immediately available to plants for growth. Total P was determined by ICP emission spectroscopy (AOAC method 985.01) after each sample was dry-ashed, treated with nitric acid and dissolved in HCl. Analysis of inorganic P consisted of a direct extraction method. A test portion of the litter sample was extracted with ammonium citrate at pH 7.0 in the presence of disodium ethylenedinitrilotetraacetic acid (EDTA) to complex calcium and magnesium.

Inorganic P was determined colorimetrically and percent P2O5 was calculated (AOAC method 993.31). Litter samples were taken on d 112 and d 136 for Exp 1 and on d 78 and d 126 for Exp 2 to coincide with pre and post- dietary nPP treatment administration. At the end of each experiment, all toms were transported to a commercial processing plant (Virginia Poultry Growers Cooperative, Hinton, VA). Both experiments were conducted according to West Virginia University Animal Care and Use Committee Guidelines (Protocol #08-0904).

 

Statistical methods: All data were statistically analyzed using the GLM procedure of Statistical Analysis System (SAS Institute, 2009). The experimental unit for all live production measurements was a pen of 76 and 90 birds for Exp 1 and 2, respectively. For the first 6 weeks of each study, data was analyzed using strain as the only factor (data not shown). After initiation of nPP treatments, each study was analyzed as a factorial. The 2 (Strain) x 2 (nPP Level) factorial randomized complete block design utilized 4 replicate pens per treatment (8 replicate pens per main effect). The main effects of Strain and nPP Level as well as the Strain x nPP Level interactions were tested. Fisher’s least significant difference multiple comparison tests were used to further compare treatment means. Linear and quadratic regression analyses were performed in order to generate prediction equations for FCR and grow-out time in Exp 2 since EW differed. If relationships were not significantly quadratic then the quadratic term was removed from the model. Prediction equations were generated by using a varying number of weekly data points due to the regression relationship being dependent upon which portion of the growth curve was utilized. Pearson correlation coefficients were also generated to describe the fit of each model. The authors do not believe that there is one correct method to generate these predictions, thus intended to provide readers with data that demonstrates this variability. Statistical significance was based on p = 0.05. Letter superscripts were used to indicate differences among means.

Experiment 1: There were no significant Strain differences in 136 d EW or mortality (Table 2). However, Strain B toms had a significant improvement in mortality adjusted FCR (2.25 vs. 2.16; p<0.022). There were no nPP Level or any Strain by nPP Level interactions for EW, FCR, or percentage mortality. There were no Strain effects for total P or inorganic P levels in the litter prior to changing dietary nPP Levels (p>0.05; Table 3) and no effects of Strain, nPP Level, or interaction on litter P at the end of the study.

Experiment 2: The Strain A toms had a 0.68 kg improvement in EW at 126 d (17.90 vs. 17.22 kg; p<0.045) but increased mortality adjusted FCR (2.21 vs. 2.18; p<0.01; Table 4). Regression equations that standardized Strain EW predicted a decreased time of production (4-5 d) and FCR (2-3 pts) for the Strain A toms (Table 5). It is important to note that Exp 2 Strain A toms were a test product and differed from Exp 1 Strain A toms. Strain did not affect percentage mortality. There was no nPP level or Strain by nPP Level interaction effects on EW, FCR or percentage mortality (p>0.05). Litter total P and inorganic P levels prior to changing dietary nPP Levels in the last 2 finisher diets were not different (p>0.05; Table 6). There was an approach to decrease in 126d litter total P (p<0.065) in toms fed the Low nPP Level diets but no effect on inorganic P and no Strain by nPP Level interactions (p>0.05; Table 6).

In both the experiments, the EW followed breeder company expected performance standards and the significant Strain differences in FCR for Exp 1 was similar to what is observed in commercial practice (Danny Wilburn, Virginia Poultry Growers Cooperative, personal communication). When designing experiments with potential commercial application, no difference in performance metrics of interest but potentially significant dietary cost savings has often been the hypothesis being tested and this was the underlying premise for the current studies. In one of the few studies with a similar objective to what was reported here, Roberson et al. (2004) reported that diets containing nPP formulated to 145% of the NRC (1994) recommendation significantly improved bone ash and other measures of tibia strength at 17 wk when compared with diets containing 100% of the NRC recommendations. The higher levels of nPP, however, also resulted in a 23% increase in litter P. The NRC (1994) recommendations for phosphorus do not include any adjustment for dietary phytase inclusion so that is an additional consideration in interpreting the experimental results. In Exp 1, the analyzed level of total P (0.55%) in the Norm treatments (Exp 1) and calculated nPP (0.37%) based on analyzed phytate concentration were similar to the NRC (1994) recommendations and also similar to the P treatment in Roberson et al. (2004) that was based on NCR recommendations. There were no differences between the Norm and Low nPP treatments in litter total P or inorganic P and the litter total P values are similar to what was reported by Roberson et al. (2004) for their NRC based P treatment. As mentioned previously, however, the diets in both the Norm and Low treatments contained phytase which would have put the actual nPP levels above the NRC (1994) or Roberson et al. (2004) NRC based P values. 

There were also no nPP Level effects on various performance metrics in Exp 2 where lower P values were incorporated in the first finisher diet and fed for a longer period of time. There was a 5 to 6% reduction in litter total P that approached significance (p<0.065) and the values were again similar to the NRC based P values reported by Roberson et al. (2004). Turner and Lilburn (1992) and Lilburn (2006) Proceedings of the Multi-State Poultry Nutrition Conference reported that linear skeletal growth is near complete by 14 week in commercial toms and the present data suggests that the latter stages of the growing period is the optimal time to reduce dietary and litter P without negatively impacting performance factors. This also presents the greatest opportunity for dietary cost savings because approximately 50-60% of total flock feed consumption is represented by the Finisher 1 and 2 diets. The Low nPP diets in both Exp 1 and Exp 2 were formulated primarily by removing inorganic rock based P. The lower levels of total P and nPP needed to further reduce litter P concentration would be dependent upon the levels of nPP coming from other dietary sources, particularly animal byproducts.

When comparing strains that vary in EW at a predetermined time, it is useful to adjust performance data to estimate FCR at a similar EW or estimate the time at which a desired EW is achieved. (Case et al., 2012, University of Guelph, Ontario, Canada). In order to better interpret Strain effect in Exp 2, linear and quadratic regression equations were generated using performance data summarized by week. However, the equations and predictions varied depending upon the derivation of the regression equation, more specifically the number of weekly performance variables used. For example, using the Exp 2, Strain B data as a basis for comparison with a 17.22 kg EW, 2.18 FCR and 126 d (18 week) grow-out time, the Strain A estimates for FCR and grow-out time differed depending on the number of weekly performance variables used to create the equation, i.e., 18 (week 1-18), 9 (week 10-18) or 3 (week 16-18) (Table 8). Using all 18 weekly ending weights, the relationship between EW and wk was quadratic (p<0.0001) and the relationship between FCR and EW was also quadratic (p<0.0001). This data set predicted that the time for Strain A toms to reach the same EW as Strain B toms would be 5 days shorter with a 3-point decrease in FCR. Using 9 weekly ending weights the relationship between EW and wk was linear (p<0.0001) and the relationship between FCR and EW was quadratic (P = 0.0098). This data set predicted that the time for Strain A toms to reach the same EW as Stain B toms would be 4 days shorter with a 2-point decrease in FCR. Using only 3 weekly ending weights the relationship between EW and wk was linear (p<0.0001) and the relationship between FCR and EW was linear (p<0.0001). This data set predicted that the time for Stain A toms to reach the same EW as Stain B toms would be 4 days shorter with a 2-point decrease in FCR. These prediction equations demonstrate that the number of data points incorporated into the regression equation may cause the predictions to vary, making it difficult to make accurate conclusions. Regardless, all prediction equations had high coefficients of determination (R2) and demonstrated that Stain A grow-out time would be reduced and FCR would be decreased relative to Strain B when ending weight was standardized. As previously mentioned, it is important to note that Strain A toms in Exp 2 were a test product and differed from Strain A toms in Exp 1.

 

 

Angel, R., N.M. Tamim, T.J. Applegate, A.S. Dhandu and L.E. Ellestad, 2002. Phytic acid chemistry: influence on phytin-phosphorus availability and phytase efficacy. J. Appl. Poult. Res., 11: 471-480.

Angel, R. and T. Applegate, 2001. Phytase use-What do we know? Pages 250-263 in Proc. 62nd Minnesota Nutrition Conf. and Minnesota Corn Growers Assoc. Technol. Symp., Bloomington, MN. Minnesota Corn Growers Assoc., Shakopee.

Case, L.A.B., J. Wood and S.P. Miller, 2012. The genetic parameters of feed efficiency and its component traits in the turkey (Meleagris gallopavo). Genetics Selection Evol., 44: 2.

Lilburn, M.S., 2006. An overview of enzyme research in turkeys and broiler breeders at Ohio State University. Multi-State Poult. Nutr. and Feeding Conf., 24-25.

National Research Council (NRC), 1994. Nutrient Requirements of Poultry. 9th rev. ed. Natl. Acad. Press, Washington, DC.

Rath, N.C., G.R. Huff, W.E. Huff and J.M. Balog, 2000. Factors regulating bone maturity and strength in poultry. Poult. Sci., 79: 1024-1032.

Roberson, K.D., M.W. Klunzinger and R.A. Charbeneau, 2004. Benefit of feeding dietary calcium and nonphytate phosphorus levels above national research council recommendations to tom turkeys in the growing-finishing phases. Poult. Sci., 83: 689-695.

SAS Institute, 2009. The SAS System for Windows 2000. Release 9.1. SAS Inst. Inc., Cary, NC.

Turner, K.A. and M.S. Lilburn, 1992. The effect of early protein restriction (zero to eight weeks) on skeletal development in turkey toms from two to eighteen weeks. Poult. Sci., 71: 1680-1686.

Van der Klis, J.D. and H.A.J. Versteegh, 1996. Phosphorus nutrition of poultry. Pages 71-83 in Proceedings of the Nottingham Feed Manufacturers Conference, Nottingham, UK.

Waldroup, P.W., 1999. Nutritional approaches to reducing phosphorus excretion by poultry. Poult. Sci., 78: 683-691.