A description of the growth of the major body components of 2 broiler chicken strains

A description of the growth of the major body components of 2 broiler chicken strains N. K. Sakomura,*^ R. M. Gous,t S. M. Marcato,t and J. B. K. Fernandes§ *Department of Animal Science, Gollege of Agrarian and Veterinary Sciences, University Estadual Paulista, Jaboticabal, Sao Paulo, Brazil, 14884-900; fDepartment of Animal Science, Faculty of Scienee and Agriculture, University of Kwazulu-Natal, Pietermaritzburg, South Africa 3209; f Department of Animal Science, Maringá State University, Maringá, Paraná, Brazil 87020; and § Aquaculture Genter of University Estadual Paulista 14884-900, Jaboticabal, Sao Paulo, Brazil ABSTRACT The objective of this work was to compare the growth and chemical composition of the main body parts of males and females of the Ross and Cobb broiler strains available in Brazil. In total, 1,920 chicks were raised in 16 floor pens, from which 4 birds of each strain and sex were sampled weekly for the required measurements. The strains and sexes were reared in separate pens, using 4 replications of 120 birds each. Four feeds, based on corn and soybeans, were used during the trial: feed 1 from 1 to 7 d, feed 2 from 8 to 28 d, feed 3 from 29 to 49 d, and feed 4 from 50 to 56 d of age (all of the birds were given the same feed for each time interval). All of the birds were weighed weekly and the 4 birds sampled from each strain and sex were weighed, fasted for 24 h, reweighed, killed, eviscerated, and dissected. The breast, drums, thighs, and wings were weighed, packed into identified plastic bags, and stored in a freezer to later be thawed and minced. They

were then freeze-dried to obtain the water content, after which they were again milled before analyzing for protein, lipid, and ash using AOAC procedures. The Gompertz equation was fitted to the weights of the body parts as well as to the protein weights of the components. Using data from each individual chicken that was sampled, allometric regressions were fitted to the weights of the physical and chemical components, with In body protein weight being the independent variable and In component weight being the dependent variable. Although some of the allometric relationships between the various body parts and body protein weight differed statistically between strains and sexes, these differences were of little commercial significance, suggesting that the relative growth rates of the different body components of the genotypes tested in this trial have not been changed substantially by genetic selection.

Key words: body part growth, allometry, breast, thigh, drumstick 2011 Poultry Science 90:2888-2896 doi:10.3382/ps.2011-01602

INTRODUCTION Commercial broilers supply a wide range of markets, which vary according to community, custom, and economic sector. Live birds are still sold in some places. A large proportion of broilers are sold dressed, with or without some of the appendages, and the most sophisticated markets demand select portions that are devoid of skin,'fät, and tíó'rle.'To cater to this wide range of tastes, the 'poultry indiis'try' has developed highly sophisticated evisceration equipment (Meyn, 2010; Stork, 2010) to produce the desired'piroduc'ts. However, little information is available with Which'to predict the weights of

©2011 Poultry Science Assoeiation Inc. Received May 12, 2011. Accepted August 8, 2011. ^ Gorrespondihg author: [email protected]

the important parts of the broiler, which vary according to age, weight, strain, sex, rearing environment, and the feeds and feeding program used during the growing period. The most comprehensive information available thus far (Hakansson et al., 1978; Knízetová et al, 1991; Hancock et al., 1995) is now outdated because of the rapid progress that has been made by geneticists in improving the performance in modern so-called high-yield genotypes (Aviagen, 2010; Cobb, 2010). The way in which an evaluation should take place is to grow male and female broilers separately, and sample birds from the population at frequent intervals such that the growth and chemical composition of the physical parts of the body can be determined and described mathematically. It is preferable to obtain the carcass protein content of the whole carcass because this may be used as the independent variable when describing allometric relationships between body components. The

2888

BODY COMPONENT GROWTH OE BROILER CHICKENS reason for using protein weight rather than BW is that body lipid, being a storage organ for excess energy, varies with different feeds and in different environments, so it should be excluded. The protein growth of the various body parts should be described using the Compertz (1825) equation, and all components for which the rate of maturing parameter (B) is the same are allometrically related, and can be predicted from the weight of another component sharing the same B value. Thus, body protein weight can be used as the independent variable, and all other allometrically related components in the body can be predicted from the weight of body protein at any stage of growth. It remains then to describe the allometric relationship between each of the components and body protein. Such an evaluation was conducted recently (Danisman and Cous, 2008, 2011), and the conclusion from that exercise was that all genotypes evaluated shared a common allometry, although small differences were apparent between broilers that had been fed widely different protein concentrations in the feed. These differences were assumed by Danisman and Cous (2008, 2011) to be because of differential amounts of lipid being deposited in the tissues, resulting from the varying dietary protein levels fed to the broilers, but no chemical analyses were conducted on those tissues to determine if this was the case. According to Fisher and Cous (2008), in the models to predict economic performance of broilers, it is necessary to include components of the body that generate revenue. These vary widely in different enterprises from the whole eviscerated carcass to dissected meat, and may include waste components in some markets, such as offal and feet. The ideal situation is to predict the growth of many such components, taking into account the bird's genotype, feeding and environmental effects, and age at slaughter. Particular problems arise from small, but commercially important differences between breeds, and from treatments at one stage of growth, such as high protein levels in starter diets, which may express modified yield components later in life. Different approaches to this problem are outlined by these authors, with the preferred method of using allometric relationships with feather-free body protein as the principle predictor, and with adjustments for body fatness and perhaps degree of maturity. The growth potential of several broiler chicken strains has been studied (Hancock et al., 1995; Cous et al., 1999; Sakomura et al., 2005; Marcato et al., 2008). However, there are few studies about growth curves of the body parts of broiler chickens. Coliomytis et al. (2003) evaluated growth of the body, breast, thigh, drumstick, wing, and breast meat of males from 2 commercial strains (Cobb 500 and Shaver Starbro). Recently, Danisman and Cous (2008, 2011) also evaluated the allometric growth of physical parts of several strains of chickens. However, information about the chemical composition of the body components of broilers during growth has not been reported previously.

2889

The objective of this work was to determine the growth parameters of major body parts, and the allometric growth of the composition and parts of female and male broiler chickens of 2 strains.

MATERIALS AND METHODS The experiment was conducted at the College of Agrarian and Veterinary Sciences, University Estadual Paulista, Jaboticabal (Sao Paulo, Brazil) for a period of 56 d. The Institution's Ethics and Animal Welfare Commission approved all procedures. In total, 1,920 sexed chicks of 2 commercial broiler strains (Ross 308 and Cobb 500) were used in the trial. The birds were individually weighed and placed into 16 pens (120 birds/pen) at a stocking density of 10 birds/ m^, and the distribution of BW was similar in each pen. A factorial arrangement was used with 2 strains and 2 sexes, each combination being replicated 4 times. During the experimental period, birds had free access to water and feed. Birds were subjected to 23L:1D throughout the trial. Chicks were brooded at a temperature of 30°C, after which the ambient temperature was reduced by 21 d of age to the prevailing temperature, which fluctuated between 13.3°C and 24.5°C, with humidity between 63.6 and 93.1%. Four corn-soybean meal diets were fed during the trial: feed 1 from 1 to 7 d, feed 2 from 8 to 28 d, feed 3 from 29 to 49 d, and feed 4 from 50 to 56 d of age (Table 1). All birds were weighed weekly, from which the average BW per pen and per treatment was calculated. Four birds representing the BW of each treatment were sampled weekly. The weights of these birds were recorded, after which they were fasted for 24 h to empty their gastrointestinal tracts. At the end of the fasting period, each bird was weighed again to determine the empty BW. They were then killed by CO2 asphyxiation, defeathered, eviscerated, and weighed once again. Then, the parts, including skin and bones, were dissected; these consisted of wings, breast, thighs, and drums. Each component was weighed, placed in a plastic bag, identified, and frozen. Upon thawing, the components were ground in an industrial meat grinder (98 BT CAF Rio Claro, Sao Paulo, Brazil) to obtain homogeneous samples, which were freeze-dried (Supermodulyo Freeze Dryer Thermo Electron Corporation, Edwards, Asheville, NC) at —50°C to determine the water content, and thereafter, ground again in the micromill (IKa A l l Basic, Ika Works do Brazil Ltda, Taquara, RJ, Brazil). These dried samples were analyzed for CP by Kjeldahl (method 954.01), ether extract (method 920.39), and ash (method 942.05), according to AOAC (1995) procedures. The mean weights and SE of means of all component weights (expressed in all cases as single components; that is thigh, drum and wing are the means of 2 appendages) and the proportions of water, protein, lipid, and ash in each of the components were determined.

2890

SAKOMURA ET AL.

Table 1. Ingredient and nutrient composition of the 4 feeds used in the trial Eeed 1 • (Prestarter; 1-7 d)

Item Ingredient (%) Corn Soybean meal Soybean oil Dicalcium phosphate Limestone Salt L-Lysine DL-Methionine 99% Mineral and vitamin supplementation Nutritional composition ME (kcal/kg) CP (%) Total lysine (%) Methionine (%) Total methionine -I- cystine {%) Calcium {%) Available phosphorous {%) Sodium (%)

55.68 36.87 3.091 2.083 0.908 0.398 0.275 0.299 0.400^ 3,010 22.0 1.38 0.63 0.97 1.00 0.50 0.20

Eeed 2 (Starter; 8-28 d) 54.18 36.36 5.457 1.825 0.950 0.400 0.132 0.291 0.400' 3,150 21.5 1.25 0.62 0.95 0.95 0.45 0.20

Feed 3 (Growth; 29-49 d) 58.20 32.53 5.536 1.847 0.835 0.353 0.064 0.233 0.4002 , 3,200 20.0 1.10 0.54 0.86 0.90 ' 0.45 0.18

Eeed 4 (Einal; 50-56 d) 64.31 26.92 5.145 1.716 0.831 0.359 0.116 0.202 0.400^ 3,245 18.0 1.00 0.48 0.78 0.85 0.42 0.18

^Mineral and vitamin supplementation (composition/kg of feed); choline, 130.5 mg; vitamin A, 5.500 MIU; vitamin D3, 1,000 MIU; vitamin E, 8.000 mg; vitamin K3, 750 mg; vitamin Bi, 600 mg; vitamin B2, 2,250 mg; vitamin B6, 1,000 mg; vitamin B12, 8.000 (xg; niacin, 17,500 mg; calcium pantothenate, 4,500 mg; folie acid, 200 mg; biotin, 25 mg; Fe, 15,000 mg; Cu, 4,500 mg; Mn, 30,000 mg; Zn, 30,000 mg; I, 500 mg; selenium, 125 mg; Bacitracin zinc, 22.500 mg; Olaquindox, 24.500 mg; and Salinomycin, 33.000 mg (Tortuga Cia Zootécnica, Mairinque, Sao Paulo, Brasil). ^Mineral and vitamin supplementation (composition/kg of feed): choline, 109.620 mg; vitamin A, 4.500 MIU; vitamin D3, 800 MIU; vitamin E, 7.000 mg; vitamin K3, 750 mg; vitamin Bi, 600 mg; vitamin B2, 2,000 mg; vitamin Bß, 900 mg; vitamin B12, 6,000 [ig; niacin, 15,000 mg; calcium pantothenate, 4,500 mg; folie acid, 150 mg; biotin, 25 mg; Fe, 15,000 mg; Cu, 4,500 mg; Mn, 30,000 mg; Zn, 30,000 mg; I, 500 mg; selenium, 125 mg; Bacitracin zinc, 22.500 mg; Olaquindox, 24.500 mg; Salinomycin, 33.000 mg (Tortuga Cia Zootécnica, Mairinque, Sao Paulo, Brasil).

The Gompertz equation was fitted to the protein weights of the body components using SAS (2001). The form of the equation used was Pt=

exp{- exp[- Bx

{t- Í*)]},

where Pt is the protein weight (g) of the component at time t; P^ is the protein weight (g) at maturity; B is the rate of maturing (g/d); and t* is the time (d) at which maximum growth rate is attained (Gompertz, 1825). Allometric regressions were fitted to both the total weight and the protein weight of each component, using the In body protein weight as the independent variable. Differences between strains and sexes were tested using a simple linear regression with groups in GenStat (2002).

RESULTS The mean weekly weights of the breast, thigh, drum, and wings, and the proportions of protein, water, lipid, and ash in these components of males and females of the 2 broiler strains used in the trial are given in Tables 2, 3, 4, and 5. The mean weekly weights of each of these components in males were always heavier than those of the females from 7 d onwards, but between strains, the weights of the components were very similar throughout. The proportion of protein in the breast was initially very low (10.7%), but increased curvilinearly (Table 6), reaching a maximum of approximately 24% before

decreasing and subsequently increasing again, the pattern being similar in both sexes and strains. There was no trend in the lipid content of the breast over time, but the water content changed linearly at a rate of —0.147%/d. The ash content remained virtually constant over the entire period. The protein and lipid contents of the thigh increased linearly throughout the growing period (Table 6), and the water content decreased hnearly over the same period. These trends were the same in both sexes and strains. However, the increase in lipid content was more consistent in the Cobb strain, with increases and decreases being evident between weeks in both males and females of the Ross strain. The ash content remained relatively constant and showed no trend over time. Whereas the protein content of the drum increased linearly over time in both sexes and strains, there were significant differences in the trends for lipid and water contents between the sexes. The rates of increase in lipid content and decrease in water content were the same in both sexes, but the intercept (Table 6) was significantly higher (1.74%) for lipid and lower (1.1%) for water in females than in males, and this applied to both strains. The protein content in the wing changed similarly to that in the breast, first increasing and then decreasing in both sexes and strains (Table 6). In this case, the changes were more consistent than those in breast protein. The water content decreased linearly with time in both sexes and strains, whereas the lipid content increased linearly in all genotypes, but the constant term for Ross males was significantly lower (2.5%) than for

BODY COMPONENT GROWTH OE BROILER CHICKENS

2891

Table 2. Mean weight (g) and proportions (%) of the chemical components of the breast and BW.(g) of males (M) and females (F) of 2 broiler strains at weekly intervals BW (g) Strain and age (d) Ross 1 7 14 21 28 35 42 49 56 Cobb 1 7 14 21 28 35 42 49 56

Breast; weight (g)

Protein {%•)

M

F

M

F

M

46.3 151 424 856 1,422 2,142 2,824 3,594 4,332

46.3 146 398 798 1,239 1,830 2,421 2,953 3,388

2.24 20.9 71.4 174 319 488 660 967 1,030

2.27 18.4 66.1 160 271 439 610 815 908

11.0 17.7 23.2 23.7 22.7 19.2 20.2 21.0 23.8

52.0 162 456 934 1,467 2,193 2,849 3,633 4,228

48.0 154 421 812 1,298 1,911 2,508 2,914 3,350

2.32 21.3 75.1 181 326 510 681 976 1,097

2.43 20.5 65.6 156 288 446 610 810 870

10.9 16.0 21.0 23.6 21.9 20.3 21.6 22.5 22.7

the other 3 genotypes. This was confirmed by using different combinations of the 4 genotypes as the reference genotype in the regression analysis. The increasing weights of protein in each component over time were used to determine the parameters for the Gompertz growth curves that described these changes. These parameters are given in Table 7 for the 4 components of the body, males and females, and the 2 strains used in the trial. The rates of maturing of the breast of males and females of the Cobb strain were faster than for Ross (0.049 vs. 0.047/d, respectively), but the mature size of Cobb (284 g) was less than that of Ross

Lipid •

F , 11.4 . .19.9 23.4 24.5 23.4 22.1 22.5 21.6 27.7 9.5 21.1 22.1 24.6 21.7 22.1 23.0 22.5 23.1

Water

Ash

{9.

(%)

M

F

8.7 9.3 10.1 5.8 6.7 5.1 5.7 7.6 7.9

8.9 9.9 10.9 5.8 8.3 6.7 7.7 10.7 11.3

9.7 7.0 9.3 5.6 7.0 7.7 6.2 9.3 10.4

8.0 9.8 9.3 7.2 6.7 8.5 7.8 9.4 10.1

•

(%)

F

M

F

79.0 71.3, 64.7 69.1 68.5 74.0 72.5 69.6 66.4

78.5 68.1 63.7 68.0 66.3 69.0 67.9 65.9 58.7

1.2 1.7 2.0 1.4 2.0 1.7 1.5 1.8 1.9

1.2 2.1 2.0 1.6 2.0 2.2 1.9 1.8 2.3

78.2 75.2 67.7 69.0 69.1 70.4 70.2 66.5 65.3

81.4 66.8 66.4 66.5 69.2 67.6 67.4 66.1 64.9

1.2 1.8 2.0 1.7 2.0 1.6 2.0 1.7 1.7

1.1 2.3 2.3 1.6 2.3 1.8 1.8 2.1 2.0

M

(345 g). The inverse relationship between mature protein weight and rate of maturation was evident with the parameters describing the growth of the thigh, but in this case, the Ross males and Cobb females exhibited greater mature weights and lower rates of maturing than the reciprocal genotypes. The rate of maturing of thighs was more rapid than that of the breast (0.049 vs. 0.044/d, respectively). The mature weight of the protein in the drum, predicted by the Gompertz equation (Table 7), was heavier for Cobb than for Ross (152 vs. 122 g, respectively), whereas the rate of maturation was faster for Ross than

Table 3. Mean weight (g) and proportions (%) of the chemical components of the thigh of males (M) and females (F) of 2 broiler strains at weekly intervals Protein

Thigh weight (l Strain and age (d) Ross 1 7 14 21 28 35 42 49 56 Cobb 1 7 14 21 28 35 42 49 56

Water

Lipid

M

F

M

F

3.53 12.0 35.9 77.4 130 203 279 350 398

3.44 10.9 32.2 66.2 109 165 222 261 287

14.7 15.0 20.1 16.5 18.4 16.2 17.4 17.7 20.5

15.7 15.0 18.5 16.3 18.0 17.8 16.3 18.1 17.3

3.75 12.4 38.3 85.4 137 210 275 . 344 388

3.65 11.4 34.5 68.7 116 173 220 285 288

14.0 14.4 18.9 16.5 18.5 17.2 16.7 16.6 18.1

14.8 15.9 17.9 17.2 19.0 18.2 17.6 19.9 20.5

M

Ash

F

M

F

M

F

6.3 13.2 4.8 10.0 5.5 11.4 11.2 16.0 13.2 •

5.1 11.3 5.1 10.6 10.6 12.2 16.7 14.0 14.9

78.3 76.5 68.5 72.0 69.7 73.6 69.8 67.7 63.3

76.9 76.5 70.7 73.3 68.4 68.2 67.4 63.6 65.6

2.7 2.6 3.4 3.1 3.8 2.6 2.6 3.7 3.1

2.6 2.7 3.1 3.2 4.7 3.2 2.6 3.4 2.3

5.6 6.0 7.1 6.5 8.5 8.0 10.9 12.0 14.1

4.2 6.4 9.9 6.5 9.6 10.9 11.6 9.9 14.1

78.1 77.0 70.5 73.9 70.2 71.7 69.5 68.7 64.8

78.7 74.9 68.2 72.6 67.9 67.6 68.1 67.3 61.6

2.3 2.7 3.5 3.1 2.9 3.1 3.0 2.7 3.0

2.3 2.7 4.0 3.8 3.6 3.2 2.6 2.9 3.8

2892

SAKOMURA ET AL.

Table 4. Mean weight (g) and proportions (%) of the chemical components of the drum of males (M) and females (F) of 2 broiler strains at weekly intervals Drum weight (g) Strain and age (d) Ross 1 7 14 21 28 35 42 49 56 Gobb 1 7 14 21 28 35 42 49 56

Protein

Lipid

Ash

M

F

M

F

M

F

M

F

4.06 13.9 40.6 89.5 160 241 354 450 508

3.84 12.5 36.9 78.7 131 196 281 368 395

•14.4 16.4 16.7 15.8 16.8 15.8 17.3 17.7 18.5

15.2 16.8 15.6 16.6 16.3 18.5 18.2 17.7 17.7

8.1 10.7 13.1 12.2 12.1 12.6 14.1 16.3 14.8

8.6 12.1 13.7 14.2 12.9 16.0 17.3 18.8 17.7

75.6 70.6 67.8 69.7 68.7 69.4 66.8 63.6 64.5

4.15 13.4 44.0 88.4 160 232 325 480 537

3.93 13.9 39.4 79.9 129 202 280 358 408

14.2 14.4 15.9 16.1 20.8 17.0 18.0 17.1 19.3

15.3 16.9 15.6 16.2 18.7 16.8 17.1 19.3 19.4

6.4 9.6 13.8 13.3 16.4 11.3 14.0 13.9 16.9

8.0 11.9 13.3 13.5 14.2 15.9 13.2 18.4 21.1

77.6 74.0 68.2 68.2 60.5 69.5 65.9 66.8 61.5

for Gobb (0.048 vs. 0.039/d, respectively). The mean rate of maturation for the drum over all genotypes was similar to that for the breast. However, the rates of maturation for the wings of the 4 genotypes were higher than for any of the other components measured (Table 7) being, on average, 0.058/d. The mature weight of protein in the wings was only 59 g, being similar for both strains, but with males having more protein at maturity than females (66 vs. 49 g, respectively). The allometric regressions fitted to the component weights and to the weight of protein in each component, using the In body protein weight as the indeTable 5. Mean weight (g) and proportion strains at weekly intervals

(%) 1

.

M

F

74.3 68.6 67.8 66.4 68.5 62.5 62.4 61.2 62.7

1.9 2.3 2.4 2.2 2.3 2.2 1.8 2.4 2.3

1.9 2.4 2.9 2.8 2.3 3.0 2.2 2.3 1.9

74.7 68.4 68.7 67.9 64.6 64.6 67.4 59.8 57.1

1.8 2.0 2.1 2.4 2.3 2.2 2.1 2.2 2.2

1.9 2.7 2.4 2.4 2.5 2.8 2.3 2.4 2.4

pendent variable are given in Table 8. The regressions for predicting thigh, drum, and wing weights were the same for all genotypes and sex. When breast weight was described in terms of body protein weight, there were differences between genotypes in both the constant term and the regression coefficient (the Gobb males differed from Ross male and females). Where the protein weights of the components were predicted from body protein weight, single equations could predict these weights in all genotypes. These allometric regressions reflect the changes in the protein content of the components relatively accurately.

of the chemiealcomponents of the wing of males (M) and females (F) of 2 broiler

Protein

Wing (g)

Strain and age (d)

Water

Lipid (%)

(%)

Water

Ash (%)

(9

M

F

M

F

M

F

M

F

M

F

1.88 10.0 32.1 64.3 107 180 219 265 310

1.90 9.41 29.3 60.5 94.9 141 186 219 245

13.8 15.3 16.1 16.2 22.0 21.4 18.7 20.7 19.4

14.0 15.5 16.7 15.3 18.1 21.9 16.7 17.8 20.4

2.1 5.1 8.3 8.6 10.8 11.1 10.5 12.2 15.5

2.9 5.5 10.2 10.8 14.6 9.8 14.6 16.9 13.9

82.5 76.8 72.4 72.9 61.6 62.3 63.3 62.1 58.3

81.5 76.2 69.3 71.4 62.6 58.1 64.7 60.3 58.3

1.6 2.7 3.3 2.6 5.5 5.1 7.5 5.0 6.8

1.6 2.8 3.8 2.6 4.8 10.2 4.0 5.1 7.3

1.94 10.3 33.1 63.9 110 166 213 258 307

1.94 10.2 31.4 59.7 90.4 142 177 216 234

11.3 13.6 16.1 16.8 18.8 19.4 17.7 17.9 18.0

11.2 15.0 17.1 16.2 19.8 18.1 18.9 16.3 17.9

4.0 5.8 9.7 11.0 7.3 13.5 13.0 14.8 12.6

3.5 7.4 9.7 9.7 12.5 13.7 15.1 17.8 17.7

83.2 77.4 70.4 69.8 68.1 62.1 65.3 62.9 65.3

83.7 74.4 69.5 70.8 61.4 64.1 61.7 60.4 60.7

1.5 3.2 3.8 2.4 5.7 5.0 4.0 4.4 4.1

1.6 3.2 3.8 3.3 6.4 4.1 4.3 5.5 3.7

Ross 1 7 14 21 28 35 42 49 56

Gobb 1 7 14 21 28 35 42 49 56

BODY COMPONENT GROWTH OF BROILER CHICKENS

2893

Table 6. Changes in the proportions (%) of the ehemieal components of the breast, thigh, drum, and vising of males (M) and females (F) of 2 broiler strains over time Constant term Item Breast Protein Lipid Water Thigh Protein Lipid Water Drum Protein Lipid (M)l Lipid (F) Water (M)l Water (F) Wing Protein Lipid Lipid^ Water

Square coefficient

Regression coefficient

Mean

SE

Mean

SE

Mean

SE

R2

13.57 8.23 73.28

1.15 0.51 1.21

0.526 0.0001 -0.147

0.096 0.016 0.036

-0.0068

0.002

57.1 NS 30.8

15.68 5.76 76.59

0.43 0.71 0.71

0.056 0.145 -0.214

0.013 0.021 0.021

33.6 56.0 73.9

15.24 8.75 10.49 74.0 72.9

0.33

0.061

0.010

52.0

0.59

0.143

0.016

72.1

0.86

-0.204

0.026

72.2

12.43 6.14 3.62 78.4

0.65

0.329

0.054

0.78

0.204

0.019

80.8

1.09

-0.372

0.033

78.5

-0.004

0.001

66.2

'^Constant terms for M and F differ. ^Constant term for Ross M differs from other strains x sex.

For example, the equation for breast weight predicts a weight of 1,045 g when BW is 3,000 g (body protein of 633 g) and the protein content in the breast is 27.7%, which compares well with the weight and composition of the breast of Ross males at 56 d of age (1,030 g and 23.8%, respectively). Similarly, at the same weight, the weight of the thigh is predicted to be 350 g with a protein content of 18%, which compares favorably with the observed thigh weight of 388 g and 18% protein content for Cobb males.

DISCUSSION Because of the different selection goals applied by geneticists over the past decades, the growth parameters of broiler genotypes may differ in several traits, including those that affect their potential growth curves.

Among these are the BW, lipid content, and feather weight at maturity, and the rates of maturing of the chemical and physical components of the body. The chemical and physical composition of the body systematically changes during growth, and an adequate description of potential growth must deal with such changes (Gous et al., 1999). Knowledge of the changes that occur in the weights of the physical parts, as well as their chemical composition, is of considerable importance when dealing with the potential growth of commercial broilers. Breast, thighs, drums, and wings are the parts of particular interest when marketing broilers, being more valuable than the rest of the body. The weights of these parts determine the revenue obtained for the bird, and will influence the optimum weight of the broiler at slaughter.

Table 7. Parameters of the Gompertz growth curve (mature protein weight, rate of maturation, and age at maximum growth rate) ± SE, describing the growth of protein in the breast, thigh, drum, and wing of males (M) and females (F) of 2 broiler strains

Item Breast Ross Cobb Thigh Ross Cobb Drum Ross Cobb Ross Cobb

M

F

Age at maximum growth rate

Rate of maturing ( B, g/d)

Mature protein weight {Pm, g)

(t*, d)

M

F

M

F

318 ± 52.6 294 ± 17.6

371 ± 45.3 272 ± 34.5

0.044 ± 0.007 0.047 ± 0.003

0.038 ± 0.004 0.049 ± 0.007

39.6 ± 4.31 36.5 ± 1.50

44.7 ± 3.46 36.4 ± 3.05

110 ± 12.5 78.4 ± 3.32

63.4 ± 5.53 85.5 ± 14.1

0.042 ± 0.005 0.052 ± 0.003

0.053 ± 0.007 0.050 ± 0.008

39.8 ± 3.04 31.9 ± 1.01

31.9 ± 2.03 36.7 ± 4.28

154 ± 24.4 183 ± 26.5

90.6 ± 4.74 120.0 ± 7.55

0.042 ± 0.006 0.037 ± 0.004

0.054 0.041

42.8 ± 4.22 • 46.7 ± 4.07

34.3 ± 1.19 40.8 ± 1.70

68.8 ±6.85 64.2 ± 3.30

51.5 ± 4.50 46.8 ± 4.86

0.058 ± 0.009 0.056 ± 0.004

0.060 ± 0.011

31.1 ± 2.19 31.6 ± 1.16

30.0 ± 1.94 28.5 ± 2.28

± 0.004 ± 0.003 0.058 ± 0.008

2894

SAKOMURA ET AL. Table 8. Parameters describing the allometric relationships between the component weights and the protein weights {In, g) of the breast, thigh, drum, and wing, and body protein weight {In, g) in males (M) and females (F) of 2 broiler strains'Constant term Item Component weight Breast^

Thigh2 Drum Wing2 Protein weight. Breast^ Thigh2 Drum Wing2

Regression coefficient

Factor

Mean

SE

Mean

SE

F Cobb male Ross male

-0.899 -0.622 -0.532 -0.883 -0.872 -0.811

0.117

0.015

0.079 0.078 0.087

1.217 1.158 1.139 1.045 1.080 0.996

-2.447 -2.903 -2.821 -2.823

0.149 0.132 • 0.129 0.145

1.230 1.105 1.118 1.044

0.028 0.025 0.024 0.027

0.015 0.015 0.016

^Where significant (F < 0.05) differences exist in either parameter between factors (strain, sex, or strain x sex) the difference (±SE) is shown. Cobb F were used as the reference group. ^One-day-old weights excluded from regression analysis.

All of the physical components of the broiler (other than feathers) have the same rate of maturation as body protein; therefore, the rates of growth of these components may be predicted by their allometric relationship wit;h body protein weight (Emmans and Fisher, 1986). These relationships are useful in understanding the changes that take place in the bird as it grows, but they are especially useful in predicting the weights of the various components during growth, and when making comparisons between genotypes (Gous, 2001). According to Fisher and Gous (2008), the linear allometric In plot of one component of the body against another provides the strongest tool for prediction of body parts. Also, because the breast has such a high economic value in some markets, the detail of how the growth of this fraction is controlled and predicted becomes very important. It has been argued (Emmans and Fisher, 1986) that the use of allometric regressions greatly simplifies the prediction of the effects of feed and environment on the weights of the chemical and physical components of the body, in that only the growth of body and feather protein need to be simulated and not each of the components individually. But these allometric regressions need to take into account the differences between genotypes and feeds if they are to be used to predict the growth of these components. The lipid content in the body, for example, is influenced by factors such as the composition of the feed (Gous et al., 1990), and the amount of lipid that is deposited in the breast, thigh, drum, and wing will differ accordingly. Therefore, allometric regressions between these parts and body protein should exclude the additional lipid that may be deposited with some feeds and not with others. Although this trial made use of only 1 feeding program (the effects of different nutrient concentrations on the relative amounts of lipid in the parts could not be measured), the changes in chemical concentration with time and between strains and sexes provide valuable informa-

tion about the changes that occur in the composition of these parts over time. When fitting the allometric regressions of breast, thigh, and wing weights against body protein content, the weights of these components at 1 d old fell below the regression fitted, thus having considerable leverage on the slope of the regression. This tendency for the breast of 1-d-old broiler chickens to be underdeveloped in relation to body protein content has been reported by Danisman and Gous (2008, 2011). Based on these observations, the data from d 1 were excluded when fitting the allometric relationships between breast, thigh, and wing weight and body protein weight. The protein content of the breast at 1 d old was only 0.48 of the content measured 2 wk later (mean over all genotypes being 10.7% at 1 d old and 22.4% at 14 d old). This compared with 0.80, 0.93, and 0.76 for the thigh, drum, and wing respectively. This corroborates the suggestion by Noy and Sklan (1998) that the embryo makes use of nutrients from the breast during the latter stages of incubation, hence the beneficial effects of in ovo feeding (Uni and Ferket, 2004), early feeding (Noy and Sklan, 2002), and the use of a high protein prestarter feed in the early feeding period (Kemp et al., 2005). It is unlikely that the protein content of the various parts would be influenced by the feeding program applied, other than that at 1 d old, as discussed above, although the rate at which this increases over time would vary with feed composition and intake. This was not the case for the lipid and water contents, which exhibited a negative correlation, especially in the thigh, drum, and wing, and which was strongly influenced by the feed composition and intake. The lipid content in the breast (excluding bones and skin) did not change over time; in this case, a negative correlation was evident between the protein and water contents. But in the other 3 parts, the lipid content increased significantly over time, thus changing the ratio between protein, water, and ash. The allometric regressions of component

BODY COMPONENT GROWTH OF BROILER CHICKENS weight and component protein weight on body protein content were expected to differ because of the variable amounts of lipid that are deposited in these components of the body, and the differences would increase with the amount of lipid deposited, although this will differ depending on the relative amounts of lipid that can be stored in each component. To understand the effect of feed composition and intake on the relative growth of these chemical components in each of the body parts, it would be necessary to determine the minimum and maximum amounts of lipid that could be deposited in each at different stages of maturity. That a single equation can be used to predict the weights of protein in each of the parts of the body under study here, irrespective of strain or sex, supports the notion that this is the better approach than including the lipid content of each component in the prediction. Differences in the weights of the 4 components were relatively greater between sexes than between strains, resulting in greater differences in the mature protein weights and rates of maturation between sexes than those between strains (Table 7). In spite of these differences, there was no difference in the allometric regressions between the 2 genotypes, implying that the mean weekly protein weights of the components were being compared at different degrees of maturity. This has important consequences: some breeding companies have heavily selected for so-called high-yield strains, in which the breast meat yield is considerably greater than in conventional strains (Aviagen, 2010; Cobb, 2010), yet when these are compared at the same body protein weight, the differences between strains disappear (Danisman and Gous, 2008, 2011). The target of selection is not only for increased breast meat yield, but also for increased body protein content at a given age, and probably a decrease in the potential to deposit body lipid. The growth parameters for breast meat obtained in the present work, for Ross and Cobb {B — 0.044 and 0.047 for males, and 0.038 and 0.049 for females; t* = 39.6 and 36.5 d for males, and 44.7 and 36.4 d for females), when compared with those described by Gous et al. (1999) for 2 strains, Ross x Arbor Acres and Steggles x Arbor Acres {B = 0.038 and 0.037 for males, and 0.035 and 0.035 for females; t* = 48.1 and 48.8 d for males, and 49.0 and 49.2 for females), suggest that breeding companies have been successful in increasing breast meat yield over the past 10 years. However, there is no proof that this has been accomplished by changing the relative proportions of the various components in the high-yield broiler. In the present work, the breast, thigh, and drum of Cobb chickens showed lower i*than that of Ross chickens, indicating the growth of these parts is faster in the Cobb strain. Genetic selection may result in differences in development patterns of breast meat, such as the Cobb 500 strain of chickens being heavier than those of the Ross 788 and Ross 288 chickens (Danisman and Gous, 2008, 2011). These authors reported differences in drum and thigh weights, and measurements

2895

of muscle for different commercial lines. Nikolova and Pavlovski (2009) also reported a higher yield of breast in the Cobb 500 strain compared with the Hubbard strain, but did not find any difference in drum and thigh weights between genotypes. The mean rate of maturation {B) and t* values of the Gompertz growth curve reported by Goliomytis et al. (2003) for breast, thigh, and drumstick weights for Cobb 500 and Shaver Starbro chickens were B — 0.038, 0.035, 0.034 /d; and t* = 47.0, 50.3, 47.4 d, respectively. In our work, the equivalent mean values over both Cobb and Ross strains for breast protein weight were B = 0.047/d for, males, and 0.052/d for females; t* = 38 d for males, and 41 d for females; and for drum, B = 0.040 /d for males, and 0.048 /d for females; t* = 44.8 d for males, and 37.6 d for females. According to these parameter values observed in our work, recent selection for these traits has achieved higher rates of maturation, thereby obtaining the maximum growth rates at an earlier age. , The allometric relationships between the various body parts and body protein weight obtained in this study were very similar between strains and sexes, even though there are statistical differences between these factors. These results corroborate those of Danisman and Gous (2008, 2011), which suggests that the relative growth rates of the different components of the body for the genotypes tested in these trials have not been substantially changed by genetic selection. In addition, these authors also studied the effect of dietary protein on the allometric relationship of body parts and concluded that the differences in the predicted weights of the physical parts when broilers are fed differing dietary protein levels may be explained on the basis that lipid is deposited to different extents in each of the parts, and it would be useful to know to what extent these rates differ, so that the weights of the parts can be adjusted by the predicted amount of lipid deposited in them on feeds differing in protein content. In spite of reports in the literature suggesting that the yield of physical parts of broiler genotypes has been altered by selection, when the weights of these parts are regressed allometrically on body protein weight, it is clear that the differences are of little economic significance when compared at the same body protein weight (Danisman and Gous, 2008, 2011).

ACKNOWLEDGMENTS The authors thank Fundaçâo de Amparo a Pesquisa do Estado de Sao Paulo - FAPESP (Brazil) for financial support of this project.

REFERENCES AOAC (Association of Official Analytical Chemists). 1995. Official Methods of Analysis. 16th ed. AOAC, Arlington, VA. Aviagen. 2010. Broiler Performanee Objectives. Accessed Sep. 2010. http : //www. aviagen. com.

2896

SAKOMURA ET AL.

Cobb. 2010. Cobb 500 Broiler Performance and Nutrition Supplement. Accessed Sep. 2010. http://www.cobb-vantress.com. Danisman, R., and R. M. Gous. 2008. Predicting the physical parts of broilers. Proc. 23rd World's Poult. Sei. J. (Suppl.). World's Poultry Science Association, Beekbergen, the Netherlands. Danisman, R., and R. M. Gous. 2011. Effect of dietary protein on the allometric relationships between some carcass portions and body protein in three broiler strains. S. Afr. J. Anim. Sei. 41:194-208. Emmans, G. C , and C. Fisher. 1986. Problems in nutritional theory. Pages 9-39 in Nutrient Requirements of Poultry and Nutritional Research. C. Fisher and K. N. Boorman, ed. Butterworths, Oxford, UK. Fisher, C , and R. M. Gous. 2008. Modelling the growth of body components in broilers. Proc. 23rd World's Poult. Sei. J. (Suppl.). World's Poultry Science Association, Beekbergen, the Netherlands. GenStat. 2002. GenStat statistical software. Release 6.1. Lawes Agricultural Trust, Rothamsted, UK. Goliomytis, M., E. Panopoulou, and E. Rogdakis. 2003. Growth curves for body weight and major component parts, feed consumption, and mortality of male broiler chickens raised to maturity. Poult. Sei. 82:1061-1068. Gompertz, B. 1825. On the nature of the function expressive of the law of human mortality and on a new method of determining the value of life contingencies. Trans. R. Phil. Sei. 115:513-585. Gous, R. M. 2001. Modelling energy and amino acid requirements in order to optimise the feeding of commercial broilers. Proc. 2nd Int. Symp. Avian Nutr. Embrapa Suinos e Aves, Concordia, Santa Catarina, Brazil. Gous, R. M., G. C. Emmans, L. A. Broadbent, and C. Eisher. 1990. Nutritional effects on the growth and fatness of broilers. Br. Poult. Sei. 31:495-505. Gous, R. M., E. T. Moran Jr., H. R. Stilborn, G. D. Bradford, and G. C. Emmans. 1999. Evaluation of the parameters needed to describe the overall growth, the ehemical growth, and the growth of feathers and breast muscles of broilers. Poult. Sei. 78:812-821. Hakansson, J., S. Eriksson, and S. A. Svensson. 1978. The influence of feed energy level on feed consumption, growth, and develop-

ment of different organs of chicks. Pages 57-59 in Report 57. Sweden Univ. Agrie. Sei., Uppsala, Sweden. Hancock, C. E., G. D. Bradford, G. C. Emmans, and R. M. Gous. 1995. The evaluation of growth parameters of six strains of commercial broiler chickens. Br. Poult. Sei. 36:247-264. Kemp, C , C. Fisher, and M. Kenny. 2005. Genotype-nutrition interactions in broilers; response to balanced protein in two commercial strains. 15th Eur. Symp. Poult. Nutr. World's Poult. Sei. Assoc, Beekbergen, the Netherlands. Knízetová, H., J. Hyánek, B. Knize, and J. Roubicek. 1991. Analysis of growth curves of fowl. I. Chickens. Br. Poult. Sei. 32:10391053. Marcato, S., N. K. Sakomura, D. Munaro, J. B. Fernandes, I. Kawauchi, and M. Bonato. 2008. Growth and body nutrient deposition of two broiler's commercial genetic lines. Braz. J. Poult. Sei. 10:117-123. Meyn. 2010. Integrated Poultry Proeessing Solutions. Accessed Oct. 2010. http://www.meyn.com. Nikolova, N., and Z. Pavlovski. 2009. Major carcass parts of broiler chicken from different genotype, sex, age, and nutrition system. Biotech. Anim. Husb. 25:1045-1054. Noy, Y., and D. Sklan. 1998. Are metabohc responses affected by early nutrition? J. Appl. Poult. Res. 7:437-451. Noy, Y., and D. Sklan. 2002. Nutrient use in chicks during the first week posthatch. Poult. Sei. 81:391-399. Sakomura, N. K., F. Longo, E. O. Rondón, C. B. V. Rabello, and A. S. Ferraudo. 2005. Modeling energy utilization and growth parameter description for broiler chickens. Poult. Sei: 84:13631369. SAS Institute. 2001. Statistical Analyses System. Version 8.2. SAS Inst. Inc., Cary, NC. Stork, 2010. Stork Poultry Processing USA. Accessed Oct. 2010. http://www.storkfoodsystems.eom. Uni, Z., and P. R. Ferket. 2004. Methods for early nutrition and their potential. World's Poult. Sei. J. 60:101-111.

Copyright of Poultry Science is the property of Poultry Science Association, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.