Effect of Protein and Energy Sources and Bulk Density of Diets on Growth Performance of Chicks 1

Transcription

1 Effect of Protein and Energy Sources and Bulk Density of Diets on Growth Performance of Chicks 1 J. L. Shelton, D. W. Dean, L. L. Southern, and T. D. Bidner 2 Department of Animal Sciences, Louisiana State University Agricultural Center, Baton Rouge, Louisiana ABSTRACT Four completely randomized designed experiments 2. In EXP 4, Diets 1 and 2 were the same as Diets 1 and (EXP) were conducted to determine the effect of energy, amino acid (AA) levels, and bulk density of diets on growth performance of chicks fed diets containing corn (C) or cornstarch:dextrose (CD) as energy sources and soybean meal (SBM) or soy protein isolate (SPI) as protein sources. The chicks were fed C-SBM diets 6 to 8 d posthatching before allotment to treatment, and the assay periods ranged from 7 to 14 d. Initial weights were 86, 93, 94, and 71 g in EXP 1 to 4, respectively. Treatments were replicated 6 to 8 times with 4, 5, or 6 chicks per replicate. In EXP 1, the treatments were 1) C- SBM (1.26% Lys and 3,200 kcal of ME/kg); 2) CD-SPI (1.26% Lys and 3,460 kcal of ME/kg); 3) CD-SPI (1.36% Lys and 3,460 kcal of ME/kg); and 4) CD-SPI (1.26% Lys and 3,200 kcal of ME/kg). The AA in all diets were increased in proportion to Lys. In EXP 2, chicks were fed C or CD as the energy source and SBM or SPI as the protein source in a 2 2 factorial arrangement. In EXP 3, the treatments were 1) C-SBM (1.26% Lys and 3,200 kcal of ME/kg); 2) CD-SPI (1.26% Lys and 3,200 kcal of ME/kg); 3) Diet 1 limit-fed to chicks consuming 90% of Diet 2; 4) Diet 2 limit-fed to chicks consuming 90% of Diet 2 in EXP 3, but fed in mash or pelleted form. In all EXP, chicks with access ad libitum to diets with SPI had decreased (P < 0.02) average daily gain (ADG) and feed intake (ADFI) compared with those with access ad libitum to diets with SBM. In EXP 2, feeding C- or CD-dextrose as the energy source had no effect (P > 0.05) on ADG, ADFI, or gain:feed. In EXP 3, ADG was decreased in chicks fed the diets with SPI relative to those fed diets with SBM, but the decrease was much greater in chicks that had access ad libitum to feed (protein source feed intake, P < 0.01) than in those limit-fed to the same feed intake. In EXP 4, ADG, ADFI, and gain:feed were decreased (P < 0.01) in chicks fed diets with SPI as the protein source. Pelleting increased (P < 0.01) ADG, ADFI, and gain:feed regardless of protein source, but the increase was much greater in chicks fed the diets with SPI (protein source feed form, P < 0.01). Results from these EXP indicate that diets with SPI compared with SBM do not result in maximum growth performance in commercial broilers, and the problem may be due somewhat to nutrient deficiency but more to the physical form of the diet. (Key words: bulk density, chicks, protein sources) 2005 Poultry Science 84: INTRODUCTION Experiments (EXP) have been conducted using soy protein isolate (SPI) in diets for chicks, but little research has been conducted to compare the growth performance of chicks fed SPI relative to those fed soybean meal (SBM) as the protein source. Shelton et al. (2003) and Batal and Parsons (2003) reported decreased growth performance in chicks fed diets with SPI relative to those fed SBM as the protein source. Shelton et al. (2003) also reported that the decrease in growth was much greater 2005 Poultry Science Association, Inc. Received for publication February 10, Accepted for publication June 15, Approved for publication by the Louisiana Agric. Exp. Sta. as Manuscript No To whom correspondence should be addressed: lsouthern@ agctr.lsu.edu. in a commercial fast-growing broiler than in a slowgrowing broiler. Shelton et al. (2003) and Batal and Parsons (2003) suggested that dietary amino acid (AA) insufficiency could have caused the decrease in growth performance, and Shelton et al. (2003) suggested that dietary energy could have caused the decrease in growth performance. However, Emmert and Baker (1995) indicated that supplementing SPI diets with Met and Thr resulted in protein quality equal to that of SBM. Their research was conducted in a slower-growing broiler identical to that used by Shelton et al. (2003) when they observed that the difference in growth performance in chicks fed SPI vs. SBM was not as great as in the commer- Abbreviation Key: ADFI = average daily feed intake; ADG = average daily gain; BD = bulk density; C = corn; CD = cornstarch/dextrose; EXP = experiment; G:F = gain:feed; npp = nonphytate phosphorus; SBM = soybean meal; SPI = soy protein isolate. 1547

2 1548 SHELTON ET AL. cial strain. In all of the aforementioned EXP, chicks fed SPI consumed less feed than those fed SBM. In the research by Shelton et al. (2003), the diets containing SPI were observed to have a lower bulk density (BD; 437 mg/cm 3 ) than the diets containing SBM (597 mg/cm 3 ). Bulk density is defined as the mass per unit volume of a material (Appel, 1994), and it is measured by weighing the amount of material needed to fill a specific volume. Feeds with low BD may result in gut fill before chicks can consume adequate feed to meet their nutrient needs. This reduced feed intake on a weight, but not volume, basis results in a reduced nutrient intake, which may be the cause of the reduced growth performance of chicks fed the SPI diets. Because of this reduced feed intake, the nutrient density in low- BD diets may need to be increased on a percentage basis. Mraz et al. (1956, 1957) indicated that diet density, along with energy content, should be used when formulating diets. However, there is a lack of research specifically determining the effect of BD of diets on growth performance in chicks. Therefore, the objective of these EXP was to determine the effect of energy and AA levels, pelleting vs. mash feed, and limit feeding vs. feeding ad libitum of diets for chicks with SPI or SBM as the protein sources and corn (C) or cornstarch:dextrose (CD) as the energy sources. MATERIALS AND METHODS The methods used in these EXP were approved by the Louisiana State University Agricultural Center Animal Care and Use Committee. Four EXP were conducted with Ross Ross commercial broilers to evaluate differences in growth performance of chicks fed diets containing C or CD as energy sources, SBM or SPI 3 as protein sources, pelleting vs. mash form of feed, and limit feeding vs. feeding ad libitum. Before the EXP began, all chicks were fed a C-SBM diet adequate in all nutrients (NRC, 1994). The chicks were held overnight without feed and water on the day before allotment to treatments. They were then weighed, wingbanded, and allotted to treatments in completely randomized designs. The chicks were housed in thermostatically controlled starter batteries with raised wire floors and continuous lighting. In EXP 1 and 2, diets with C or SBM were formulated using mineral and AA values from NRC (1994). In all EXP, the diets using SPI were formulated using analyzed mineral and AA values for SPI (Table 1). In EXP 3, diets with SPI or SBM were formulated using analyzed values for minerals and AA (Table 1). The AA composition was determined after acid hydrolysis (AOAC, 2000; Method E(a)). Total sulfur AA content was determined after performic acid oxidation followed by acid hydrolysis [AOAC, 2000; Method E(b)]. Tryptophan con- 3 John R. White Co., Inc., Birmingham, AL. 4 Model Optima 3000, Perkin Elmer, Norwalk, CT. tent was determined after alkaline hydrolysis (AOAC, 2000; Method E(c)). Amino acid analyses were determined by HPLC with postcolumn ninhydrin detection. The mineral composition was determined after digestion in nitric acid and hydrogen peroxide by inductively coupled plasma emission spectroscopy (AOAC, 1990). 4 Diets (Tables 2 and 3) were formulated to meet or exceed the AA, mineral, and vitamin requirements of chicks 0 to 21 d posthatching (NRC, 1994). With the exception of EXP 4, all diets were fed in mash form. At the end of each EXP (d 20, 21, 14, and 17 for EXP 1 to 4, respectively), all chicks were weighed individually and pen feed intake was measured. Growth performance was evaluated by average daily gain (ADG), average daily feed intake (ADFI) on a weight and volume basis, and gain:feed (G:F) on a weight and volume basis. Average daily feed intake on a volume basis was determined by the following equation: (ADFI, g/bd, mg/ cm 3 ). Gain:feed on a volume basis was determined by the following equation: (ADG, g/adfi, cm 3 ). Bulk density was determined by filling a glass cube of known volume with each diet until overfilled. Excess diet was removed by sliding a plastic ruler over the rim of the cube without compressing the feed. The weights of the cube and contents were recorded and BD was calculated (mg/cm 3 ). Each value was an average of 3 independent measurements. Experiment 1 was conducted to evaluate ME and Lys levels in a CD-SPI diet relative to a C-SBM diet on growth performance of chicks. On d 6 posthatching, 192 male chicks were allotted to 4 treatments with 8 replications of 6 chicks per replicate pen. Average initial and final BW were 86 and 515 g, and the EXP lasted 14 d. The treatments were 1) C-SBM diet (1.26% Lys and 3,200 kcal of ME/kg), 2) CD-SPI control diet (1.26% Lys and 3,460 kcal of ME/kg) previously used in our lab (Shelton et al., 2003), 3) CD-SPI diet with a ME:Lys ratio equal to that in Diet 1 by increasing the Lys level to 1.36%, and 4) CD-SPI diet with a ME:Lys ratio equal to that in Diet 1 by decreasing the ME level to 3,200 kcal/ kg by adding cellulose. Chicks had access ad libitum to feed and water throughout the EXP. Diets were formulated to contain 1.00% Ca and 0.45% nonphytate P (npp). In all diets, AA met or exceeded their ratio to Lys according to NRC (1994). Experiment 2 was conducted to evaluate the protein sources (SBM or SPI) and the energy sources (C or CD) used in EXP 1 to determine whether the results observed inexp1wereduetospiorcd.ond8posthatching, 192 female chicks were allotted to 4 treatments with 8 replications of 6 chicks per replicate pen. Average initial and final BW were 93 and 423 g, and the EXP lasted 14 d. The treatments were 1) C-SBM diet, 2) C-SPI diet, 3) CD-SBM diet, and 4) CD-SPI diet. All diets were formulated to contain 3,200 kcal/kg, 1.26% Lys, 1.00% Ca, and 0.45% npp. Chicks had access ad libitum to feed and water throughout the EXP. Experiment 3 was conducted to evaluate the effect of limit-feeding vs. ad libitum feeding of the diets con-

3 SOY PROTEIN SOURCE, BULK DENSITY, AND DIET FORM 1549 TABLE 1. Analyzed mineral and amino acid content of ingredients 1 Experiments 1 and 2 Experiment 3 Experiment 4 SPI Corn SBM SPI Corn SBM SPI Thr, % Met, % Cys, % Ile, % Phe, % His, % Lys, % Arg, % Trp, % Val, % Ca, % P, % Fe, ppm Zn, ppm Cu, ppm Na, % Mg, % Mn, ppm K, % SBM = soybean meal; SPI = soy protein isolate. taining SBM or SPI as the protein source and C as the energy source on the growth performance of chicks. Average initial and final BW were 94 and 272 g and the EXP lasted 7 d. The treatments were 1) C-SBM diet fed ad libitum, 2) C-SPI diet fed ad libitum, 3) C-SBM diet limit fed, and 4) C-SPI diet limit fed. The chicks that were limit-fed were fed 90% of the intake of chicks fed the C-SPI diet. All diets were formulated to contain 3,200 kcal of ME/kg, 1.26% Lys, 1.00% Ca, and 0.45% npp. On d 6 posthatching after a 16-h fast, 128 female chicks were allotted, wingbanded, and placed in pens. There were 8 replications of 4 chicks per replicate pen. The chicks with access ad libitum to feed were fed their respective diets. Based on previous research in our lab (Shelton et al., 2003), 7-d-old chicks eat approximately 21% of their BW daily. Therefore, the chicks that were limit-fed were fed 15% of their BW on d 6 to make sure all feed was consumed. For all other days, feed intake was determined each morning on chicks with access ad libitum to the C-SPI diet because data from EXP 2 indicated that chicks fed this diet had reduced feed intake relative to those fed a C-SBM diet. The feed allowance was divided into 3 equal amounts, and chicks were fed at 600, 1400, and 2200 h. Throughout the EXP, all chicks were allowed access ad libitum to water. Experiment 4 was conducted to determine the effect of form of feed (pelleted vs. mash) in chicks fed C-SBM or C-SPI diets. On d-4 posthatching, 120 female chicks were allotted to 4 treatments with 6 replications of 5 chicks per replicate pen. Average initial and final BW were 71 and 538 g, and the EXP lasted 14 d. The treatments were 1) C-SBM diet fed in mash form, 2) C-SPI diet fed in mash form, 3) C-SBM diet fed in pellet form, 5 Model CL-3 laboratory pellet mill, California Pellet Mill Co., San Francisco, CA. and 4) C-SPI diet fed in pellet form. All diets were formulated to contain 3,200 kcal/kg, 1.26% Lys, 1.00% Ca, and 0.45% npp. The pellet mill 5 had a high chrome die with a 0.48 cm hole diameter and the die was 3.81 cm thick. Chicks had access ad libitum to feed and water throughout the EXP. Statistical Analysis Data were analyzed by ANOVA procedures appropriate for completely randomized designs (Steel and Torrie, 1980). In EXP 1, treatment means were compared with the PDIFF option in SAS (SAS, 2003). In EXP 2, 3, and 4, orthogonal contrasts appropriate for a 2 2 factorial arrangement of treatments were used to determine treatment effects. In all EXP, pen of chicks was the experimental unit for all data. RESULTS In EXP 1, ADG, ADFI (g), and G:F (g/cm 3 ) were decreased (P < 0.05) in chicks fed all CD-SPI diets relative to those fed the C-SBM diet (Table 4). Daily gain was increased (P < 0.05) in chicks fed the CD-SPI diet with 1.36% Lys and 3,460 kcal/kg relative to those fed the CD-SPI with 1.26% Lys and 3,460 kcal/kg. Gain:feed (g/g) of chicks fed the CD-SPI diet with 1.26% Lys and 3,200 kcal/kg was not different (P > 0.05) from chicks fed the C-SBM diet but was increased (P < 0.05) relative to those fed the CD-SPI diet with 1.26% Lys and 3,460 kcal/kg. Daily feed intake (cm 3 ) was decreased (P < 0.05) in chicks fed the CD-SPI with 1.26% Lys and 3,460 kcal/ kg or the CD-SPI diet with 1.26% Lys and 3,200 kcal/ kg relative to those fed the C-SBM diet. Gain:feed (g/ g) tended to be increased (P < 0.08) in chicks fed the CD-SPI diet with 1.36% Lys relative to those fed the C- SBM diet. Gain:feed (g/cm 3 ) tended to be increased (P

4 1550 SHELTON ET AL. TABLE 2. Diet composition for EXP 1 1 Item C-SBM CD-SPI CD-SPI CD-SPI ME, kcal/kg 3,200 3,460 3,460 43,200 Lys, % ME:Lys, kcal/mg Ingredient Corn Cornstarch Dextrose SBM, 47.5% CP SPI, 85.5% CP Soy oil Cellulose Monocalcium phosphate Limestone Sodium chloride Mineral premix Vitamin premix Se premix Sodium chloride DL-methionine KHCO MgSO 4 7H 2 O CuSO 4 5H 2 O L-Thr CaCl Glycine Calculated composition CP, % Crude fiber, % Sulfur AA, % Met, % Thr, % Trp, % Arg, % Ca, % npp, % Mg, ppm 1, K, % Na, % Cl, % Cu, ppm Zn, ppm Mn, ppm Fe, ppm Bulk density 6,g/cm EXP = experiment; C = corn; CD = cornstarch:dextrose; SBM = soybean meal; SPI = soy protein isolate; npp = nonphytate P. 2 Provided per kilogram of diet: vitamin A (retinyl palmitate), 1,366 IU; vitamin D 3 (cholecalciferol), 450 IU; vitamin E (DL-α-tocopheryl acetate), 50 IU; menadione (menadione sodium bisulfite), 1.5 mg; vitamin B 12, 0.02 mg; d-biotin, 0.6 mg; folacin (folic acid), 6 mg; niacin, 50 mg; d-pantothenic acid, 18.3 mg; pyridoxine (pyridoxine HCl), 6.4 mg; riboflavin, 15 mg; thiamin (thiamin HCl), 13.4 mg. 3 Provided per kilogram of diet: copper (copper sulfate 5H 2 O), 4 mg; iodine (potassium iodate), 1.0 mg; iron (ferrous sulfate 7H 2 O), 60 mg; manganese (manganese sulfate H 2 O), 60 mg; selenium (sodium selenite), 0.1 mg; zinc (zinc sulfate 7H 2 O), 44 mg; calcium (calcium carbonate), 723 mg. 4 Sodium selenite, Prince Agri Products Inc., Quincy, IL. 5 Contains 750,000 mg/kg of choline. 6 Bulk density is based on actual analysis of 3 samples. The SD was less than for all samples. < 0.06) in chicks fed the CD-SPI diet with 1.36% Lys and 3,460 kcal/kg relative to those fed the CD-SPI diet with 1.26% Lys and 3,460 kcal/kg. In EXP 2, ADG, ADFI (g), and G:F (g/g and g/cm 3 ) were decreased (P < 0.02) in chicks fed SPI compared with those fed SBM (Table 5). Daily feed intake (cm 3 ) was increased (P < 0.02) and G:F (g/cm 3 ) was decreased (P < 0.02) in chicks fed diets with CD as the energy source relative to chicks fed the diets with C as the energy source. Gain:feed (g/g and g/cm 3 ) was decreased in chicks fed the diets with SPI as the protein source regardless of the energy source used, but the decrease was greater in chicks fed the diet using C relative to CD as the energy source (protein source energy source, P < 0.04). In EXP 3, ADG, ADFI (g), and G:F (g/g and g/cm 3 ) were decreased (P < 0.01), and ADFI (cm 3 ) was increased (P < 0.01) in chicks fed the diets with SPI as the protein

5 SOY PROTEIN SOURCE, BULK DENSITY, AND DIET FORM 1551 TABLE 3. Diet compositions for EXP 2, 3, and 4 1 Experiment 2 Experiment 3 Experiment 4 Ingredient C-SBM C-SPI CD-SBM CD-SPI C-SBM C-SPI C-SBM C-SPI Corn Cornstarch Dextrose SBM, 47.5% CP SPI, 85.5% CP Soy oil Cellulose Monocalcium phosphate Limestone Sodium chloride Mineral premix Vitamin premix Se premix Choline chloride DL-Methionine KHCO MgSO 4 7H 2 O CuSO 4 5H 2 O L-Thr ZnSO 4 7H 2 O MnSO 4 7H 2 O FeSO 4 7H 2 O CaCl 0.45 Glycine Calculated composition ME, kcal/kg 3,200 3,200 3,200 3,200 3,200 3,200 3,200 3,200 ME:Lys, kcal/mg CP, % Crude fiber, % Lys, % Sulfur AA, % Met, % Thr, % Arg, % Ca, % npp, % Mg, ppm 1, , , ,648 1,648 1,606 1,606 K, % Na, % Cl, % Cu, ppm Zn, ppm Mn, ppm Fe, ppm Bulk density 6,g/cm EXP = experiment; C = corn; SBM = soybean meal; SPI = soy protein isolate; CD = cornstarch:dextrose; npp = nonphytate P. 2 Provided per kilogram of diet: vitamin A (retinyl palmitate), 1,366 IU; vitamin D 3 (cholecalciferol), 450 IU; vitamin E (DL-α-tocopheryl acetate), 50 IU; menadione (menadione sodium bisulfite), 1.5 mg; vitamin B 12, 0.02 mg; d-biotin, 0.6 mg; folacin (folic acid), 6 mg; niacin, 50 mg; d-pantothenic acid, 18.3 mg; pyridoxine (pyridoxine HCl), 6.4 mg; riboflavin, 15 mg; thiamin (thiamin HCl), 13.4 mg. 3 Provided per kilogram of diet: copper (copper sulfate 5H 2 O), 4 mg; iodine (potassium iodate), 1.0 mg; iron (ferrous sulfate 7H 2 O), 60 mg; manganese (manganese sulfate H 2 O), 60 mg; selenium (sodium selenite), 0.1 mg; zinc (zinc sulfate 7H 2 O), 44 mg; calcium (calcium carbonate), 723 mg. 4 Sodium selenite, Prince Agri Products Inc., Quincy, IL. 5 Contains 750,000 mg/kg of choline. 6 Bulk density is based on actual analysis of 3 samples. The SD was less than for each sample. source relative to those fed the diets with SBM as the protein source (Table 6). Daily gain, ADFI (g and cm 3 ), and G:F (g/g and g/cm 3 ) were decreased (P < 0.01) in limit-fed chicks relative to those with access ad libitum to the diets. Daily gain was decreased in chicks fed the diets with SPI as the protein source, but the decrease was much greater in chicks that had ad libitum access to the diets compared with those that were limit-fed (protein source feed intake, P < 0.01). Daily feed intake (g) was decreased in chicks that had ad libitum access to the diets with SPI as the protein source but not affected in chicks that were limit-fed the diets with SPI as the protein source (protein source feed intake, P < 0.01). Daily feed intake (cm 3 ) was increased in chicks

6 1552 SHELTON ET AL. TABLE 4. Growth performance of chicks fed a C-SBM diet or CD-SPI diets with an increased amino acid or energy level in EXP 1 1 C-SBM CD-SPI CD-SPI CD-SPI ME, kcal/kg: 3,200 3,460 3,460 3,200 Lys, %: Item ME:Lys, kcal/mg: SEM Daily gain, g a c b b,c 0.72 Daily feed intake, g a b b b 0.68 Gain:feed, g/g a b a a Daily feed intake, cm a b a,b b 1.41 Gain:feed, g/cm a b b b a,b,c Means within a row with different superscripts differ, P < Data are means of 8 replications of 6 chicks (6 to 20 d posthatching) per replication with an initial and final BW of 86 and 516 g. EXP = experiment; C-SBM = corn-soybean meal; CD-SPI = cornstarch:dextrose-soy protein isolate. Treatment means were compared with the PDIFF option in SAS (SAS Inst., Inc., Cary, NC). fed the diets with SPI as the protein source at either feeding level, but the increase was greater in chicks that were limit-fed (protein source feed intake, P < 0.01). In EXP 4, ADG, ADFI, and G:F were decreased (P < 0.01) in chicks fed diets with SPI as the protein source (Table 7). Also, chicks fed the pelleted diets had an increased (P < 0.01) ADG, ADFI, and G:F compared with chicks fed the mash diets, but the increase was much greater in chicks fed the diets with SPI than in those fed SBM (protein source feed form, P < 0.01). DISCUSSION The processing of SBM to soy protein concentrate involves treatment with hexane followed by an alcohol or water wash, which results in the removal of oligosaccharides (Beery, 1989). Soy protein isolate is obtained by further processing with alkaline extraction to remove cotyledons followed by acid precipitation, after which soy whey is removed (Johnson and Kikuchi, 1989). Soy whey is rich in both sulfur AA and Thr and removing soy whey makes sulfur AA and Thr the first 2 limiting AA in diets with SPI as the protein source (Emmert and Baker, 1995). Also, processing SBM to SPI greatly decreases the particle size, resulting in a product with a low BD. In our EXP, growth performance was decreased in chicks fed diets with SPI relative to those fed SBM as the protein source. This agrees with the results of Welch et al. (1988), Shelton et al. (2003), and Batal and Parsons (2003), who reported that chicks fed SPI relative to SBM as the protein source had reduced growth performance due to a reduction in ADFI. Emmert and Baker (1995) indicated that supplementing SPI diets with Met and Thr resulted in protein quality equal to that of SBM, but chicks fed SPI gained 16.5 g less and ate 50 g less relative to those fed SBM. Batal and Parsons (2003) also suggested that the decrease in growth performance of chicks fed the diets with SPI may have been caused by an AA imbalance or by the low sulfur AA and Thr concentrations that are found in processed soy products. However, in the study by Batal and Parsons (2003), the diets with SBM or SPI had equal amounts of Lys and sulfur AA, and the AA levels in the diets were at or above those recommended by NRC (1994) for 0- to 21-d-old chicks. Furthermore, they indicated that the availabilities of Lys, sulfur AA, and Thr were greater in SPI than in SBM. In our studies, dietary Lys, sulfur AA, and Thr levels were all at or above the requirement (NRC, 1994) for 0- to 21-d-old chicks and growth performance was decreased in chicks fed the SPI diets relative to chicks fed the diets with SBM. Increasing the AA in the CD- TABLE 5. Effect of protein and energy source on growth performance of chicks in EXP 2 1,2 Item C-SBM C-SPI CD-SBM CD-SPI SEM Daily gain, g a Daily feed intake, g a Gain:feed, g/g a,b Daily feed intake, cm 3c Gain:feed, g/cm 3 a,b,c a Protein source, P < b Protein source energy source, P < c Energy source, P < Data are means of 8 replications of 6 chicks (7 to 21 d posthatching) per replication with an initial and final BW of 93 and 470 g. EXP = experiment; C = corn; CD = cornstarch:dextrose; SBM = soybean meal; SPI = soy protein isolate. Orthogonal contrasts appropriate for a 2 2 factorial arrangement of treatments were used to compare the protein source and energy source effects and the pen of chicks was the experimental unit for all data. 2 All diets contained the same levels of ME and limiting essential AA.

7 SOY PROTEIN SOURCE, BULK DENSITY, AND DIET FORM 1553 TABLE 6. Effect of protein source and bulk density on growth performance of chicks in EXP 3 1,2 Diet: C-SBM C-SPI C-SBM C-SPI Item Feeding level: Ad libitum Ad libitum Limit-fed Limit-fed SEM Daily gain, g a,b,c Daily feed intake, g a,b,c Gain:feed, g/g a,b Daily feed intake, cm 3 a,b,c Gain:feed, g/cm 3 a,b a Protein source, P < b Feed intake, P < c Protein source feed intake, P < Data are means of 6 replications of 4 chicks (7 to 14 d posthatching) per replication with an initial and final BW of 94 and 272 g. EXP = experiment; C = corn; CD = cornstarch:dextrose; SBM = soybean meal; SPI = soy protein isolate. Orthogonal contrasts appropriate for a 2 2 factorial arrangement of treatments were used to compare the protein source and feeding level effects and the pen of chicks was the experimental unit for all data. 2 All diets contained the same levels of ME and limiting essential AA. SPI in EXP 1 improved ADG and G:F relative to those fed the CD-SPI with 1.26% Lys, but not to the level of those fed the C-SBM diet. This response suggests that the AA levels were not the main cause of the reduced growth performance of chicks fed the SPI diets, but that the C-SPI with 1.26% Lys diet may have been limiting in one or more AA. This improved performance from AA lends support to the suggestion of Batal and Parsons (2003) that the diet may have been limiting in an AA. It is difficult to explain why a diet that seems to provide available AA in excess of the requirement will still respond to AA supplementation. In EXP 3, the diets were formulated to contain the same amount of AA and ME. Also, C was used as the energy source because in EXP 2, chicks fed C or CD had equal growth performance, indicating that feeding CD as the energy source was not the cause of the reduced growth performance in chicks fed the CD-SPI diet in these EXP and in those of Shelton et al. (2003). In EXP 3, feeding chicks the same amount of C-SBM or C-SPI resulted in growth performance much more similar than when chicks had access ad libitum to the diets. This response indicates that feed intake is the major factor limiting growth in chicks fed SPI diets relative to those fed SBM diets. This reduction in feed intake may be due to physical form (BD or dustiness) of the diet or to possible retained chemical in SPI after processing. Feed intake on a volume basis in chicks fed SPI is similar to or greater than that of chicks fed SBM. However, ADFI on a weight basis always is reduced in chicks fed SPI. Bulk density of diets with SPI is lower than diets containing SBM. This dietary BD difference and similar ADFI on a volume basis suggests that BD may be one factor that limits nutrient and ADFI (g) in chicks fed SPI. However, in our EXP 3, ADG was decreased in the limit-fed chicks fed the SPI diet relative to those fed the SBM diet, which also indicates a nutritional problem with the SPI diet. Limit-fed chicks fed the SPI diet wasted a small amount of feed relative to those limit-fed the SBM diet, which can be seen in the ADFI data (ADFI of 0.43 g per d less when chicks were limit-fed the SPI diet). This small reduction in ADFI could have caused a portion of the reduction in ADG observed in chicks that were limit-fed SPI relative to SBM as the protein source. Also, in EXP 2, increasing the AA level in the SPI resulted in increases in ADG and G:F. Thus, the C- SPI diet may be limiting in one or more AA. In EXP 4, broilers fed SPI had a reduced ADFI compared with those fed SBM, but the decrease in feed intake was 19% in the mash diets and 12% in the pelleted diets. Thus, pelleting the diets overcame a portion, but not all of the decrease in ADFI observed in chicks fed the diets containing SPI, even though pelleting changed the BD from 0.50 to 0.62 g/cm 3. It is possible that as the C- SPI pellet is broken down in the crop and exposed to moisture it may expand more than the C-SBM pellets, thus decreasing the capacity for intake. TABLE 7. Growth performance of chicks from 4 to 17 d posthatching in experiment 4 a C-SBM C-SPI C-SBM C-SPI Item Mash Mash Pellet Pellet SEM Daily gain, g b,c,d Daily feed intake, g b,c Gain:feed, g/g b,c,d Bulk density, g/cm a Data are means of 6 replications of 5 birds each with an initial and final BW of 71 and 538 g. b Protein source, P < c Feed form, P < d Protein source feed form, P < 0.01.

8 1554 SHELTON ET AL. These data indicate that diets containing SPI do not result in maximum growth performance in commercial broiler chicks, due primarily to a decrease in ADFI. This decrease seems to be due partly to a nutrient deficiency, but more to physical form of the diet. The decrease in BD of diets with SPI may limit ADFI on a weight basis such that nutrient intake is reduced. Other factors that may limit feed intake of diets containing SPI are reduced particle size and increased dustiness, but pelleting did not completely overcome these potential problems. It seems that the reduced ADFI by broilers in diets containing SPI may be due to a myriad of factors including nutrient deficiency, BD, and/or particle size. ACKNOWLEDGMENTS The authors thank R. Payne, B. Perkins, J. Bingham, E. Savoy, and T. O Connor-Dennie (Louisiana State University, Baton Rouge) for their assistance with data collection. REFERENCES AOAC Official Methods of Analysis. 15th ed. Assoc. Off. Anal. Chem., Arlington, VA. AOAC Official Methods of Analysis. 17th ed. Official Method (a,b,c). AOAC Int., Gaithersburg, MD. Appel, W. B Physical properties of feed ingredients. Page 551 in Feed Manufacturing Technology. R. R. McEllhiney, ed. Am. Feed Ind. Assoc., Arlington, VA. Batal, A. B., and C. M. Parsons Utilization of different soy products as affected by age in chicks. Poult. Sci. 82: Beery, K. E Preparation of soy protein concentrate products and their application in food systems. Pages in Proc. World Cong. Veget. Protein Util. Human Foods and Anim. Feedstuffs. Am. Oil Chem. Soc., Champaign, IL. Emmert, J. L., and D. H. Baker Protein quality assessment of soy products. Nutr. Res. 15: Johnson, D. W., and S. Kikuchi Processing for producing soy protein isolates. Pages in Proceedings of the World Congress on Vegetable Protein Utilization in Human Foods and Animal Feedstuffs. Am. Oil Chem. Soc., Champaign, IL. Mraz, F. R., R. V. Boucher, and M. G. McCartney The influence of dietary productive energy and fiber on growth response in chickens. Poult. Sci. 35: Mraz, F. R., R. V. Boucher, and M. G. McCartney The influence of the energy:volume ratio on growth response in chickens. Poult. Sci. 36: NRC Nutrient Requirements of Poultry. 9th rev. ed. Natl. Acad. Press, Washington, DC. SAS SAS 9.1 (TS1M2). SAS Inst., Inc., Cary, NC. Shelton, J. L., I. Mavromichalis, R. L. Payne, L. L. Southern, and D. H. Baker Growth performance of different breed crosses of chicks fed diets with different protein and energy sources. Poult. Sci. 82: Steel, R. G. D., and J. H. Torrie Principles and Procedures of Statistics: A Biometrical Approach. 2nd ed. McGraw-Hill Book Co., New York, NY. Welch, C. C., C. M. Parsons, and D. H. Baker Further investigation of the dietary protein level-monensin interrelationship in broiler chicks: Influence of dietary protein source and type of anticcoccicial drug. Poult. Sci. 67: