Broiler Meat Quality: Effect of Selection for Increased Carcass Quality ...

Broiler Meat Quality: Effect of Selection for Increased Carcass Quality and Estimates of Genetic Parameters E. LE BIHAN-DUVAL,*,1 N. MILLET,* and H. REMIGNON† *Institut National de la Recherche Agronomique, Station de Recherches Avicoles, 37380 Nouzilly, France, and †Ecole Nationale Supe´rieure Agronomique de Toulouse, BP 107, Auzeville Tolosane, 31326 Castanet-Tolosan Ce´dex, France ABSTRACT Ultimate pH (pHu), color measurements, and water holding capacity of the chicken Pectoralis major muscle were compared between birds of an experimental line selected for 13 generations for increased BW and breast meat yield and reduced abdominal fat percentage and its control line. Ultimate pH differed slightly between lines after selection, with values of 5.78 ± 0.10 and 5.68 ± 0.12 in the selected and control birds, respectively. Drip loss was significantly lower in the selected birds. Although selection did not modify lightness L*, it led to paler meat, as redness a* and yellowness b* were significantly lower in the selected line than in the control line. Ultimate pH of the meat was related to lightness and drip loss (with mean

correlations over both lines of –0.59 and –0.40, respectively). Storage of the meat resulted in similar color variation in both lines, with a significant increase in a* and b* until 3 d postslaughter and in L* after 6 d postslaughter. Estimates of the genetic parameters of the criteria of meat quality were calculated in the selected line. The estimates suggested that there is a predominant role of genetics in the control of these traits, with heritability estimates of 0.49 ± 0.11 for pHu, 0.75 ± 0.08 for L*, 0.81 ± 0.04 for a*, and 0.64 ± 0.06 for b*. A significant negative genetic correlation (–0.65) was found between pHu and L*. The genetic correlation between a* and b* measurements was estimated at 0.72.

(Key words: broiler, meat quality, genetic parameters, pH, color) 1999 Poultry Science 78:822–826

Although selection of meat-type chickens for increased breast meat yield has been successful, the impact on meat quality, and especially on storage and processing quality, remains to be clarified. The main characteristics of the meat to be considered are pH, color, and water holding capacity. In this study, these traits were compared in an experimental line selected for increased breast and reduced abdominal fat and its control line (Ricard et al., 1994; Le Bihan-Duval et al., 1998). Estimates of heritability of these characteristics and of genetic correlations with growth and body composition in broiler chickens are also reported.

INTRODUCTION Selection of meat-type chickens has previously focused not only on increased growth performance but also on improved carcass quality. In particular, the emphasis has been on better body composition, with higher breast meat yield and lower abdominal fat. This focus responds to the consumer desire for healthier meat, and to the evolution of the market through a rising demand for portioned and processed products (Barton, 1994). The profitability of broiler production is therefore largely determined by the possibility of increasing the proportion of prime parts in the carcass, mainly breast meat, and by reducing fat. Body composition can be significantly improved by selection, as shown by the high level of the heritability of the amount of meat, ranging from 0.40 and 0.65 in the studies of Vereijken (1992), Jego et al. (1995), and Le Bihan-Duval et al. (1998). For abdominal fat, heritability ranges between 0.50 and 0.80 (Chambers, 1990).

MATERIALS AND METHODS

Animals The animals originated from the Quality line, selected for 13 generations for increased BW and breast meat yield (BRY) and decreased percentage of abdominal fat (FTP), and its unselected control line. After 12 generations of

Received for publication April 7, 1998. Accepted for publication January 16, 1999. 1To whom correspondence should be addressed: lebihan@tours. inra.fr

Abbreviation Key: BRY = breast meat yield; DL = drip loss; FTP = abdominal fat percentage; pHu = ultimate pH.

822

SELECTION OF BROILERS AND MEAT QUALITY

selection, differences between the two lines for BW, BRY, and FTP, respectively, were at 0.7, 1.4, and 1.0 phenotypic standard deviations (Le Bihan-Duval et al., 1998). At each generation of selection a special hatch of animals of the Quality line was reared and dissected at 6 wk of age to measure body composition traits. Additionally, the meat quality of the Pectoralis major muscle was evaluated in the 13th generation by measuring ultimate pH (pHu) and color values 24 h postmortem. These data, which were collected for a total of 231 birds of both sexes originating from 16 sires and 50 dams, were used to estimate genetic parameters of these meat quality traits. A sample of 66 males was selected from these birds for additional measurements of color stability and drip loss (DL) during storage of the meat. The same measurements were performed in parallel on 65 male birds of the control line, in order to compare meat quality traits between lines. Male birds in each line were chosen within each half-sib family (about four males per sire), so that the weights of the retained birds were the closest to the mean weight of the whole male population. The animals used for this study were reared under similar conditions in a conventional poultry house of the Avian Research Centre (INRA-SRA) in Nouzilly, France. Feed and water were provided for ad libitum consumption throughout the growth period. Birds were weighed and slaughtered after 12-h feed withdrawal in a conventional processing plant at 6 wk of age. The whole carcasses were stored for 1 d at 4 C and then dissected to measure body composition and meat quality parameters.

Meat Quality Parameters Ultimate pH was measured 24 h postmortem with a portable pH-meter2 equipped with a xerolyte electrode. Measurements were performed by placing the electrode onto the outer side of the cranial third of the right P. major muscle. The pH meter was calibrated by measuring buffer solutions (pH = 4 and pH = 7) after every 50 observations. Drip loss was measured as described by Remignon et al. (1996). The left P. major muscle was excised and weighed, then placed in a polyethylene bag and stored at 4 C. The muscle was removed from the bag 3, 6, and 10 d postslaughter, wiped, and weighed to evaluate DL, which was expressed as a percentage of the initial muscle weight. Color measurements were performed on the ventral side of the cranial third of the left P. major muscle using a Miniscan spectrocolorimeter3 with the CIELAB (L*, a*, b*) system. The latter describes opponent-color scales, based on the opponent-color theory of human color vision. a* (redness) indicates redness when positive and greeness when negative. b* (yellowness) indicates yellowness when positive or blueness when negative. L* (lightness) describes the relationship between reflected and absorbed

2Model 506, Crison 3Hunterlab, Reston,

Instruments, SA, Spain. VA 20190.

823

light, with value of 100 for white and of 0 for black. These measures can thus be used to evaluate variations in meat color, depending on the concentration and chemical state of meat pigments and of the physical characteristics of the meat, such as its light scattering and absorbing properties. Color was first measured 24 h postmortem. Further measurements were performed at 3, 6, and 10 d postslaughter after storage at 4 C in a polyethylene bag to evaluate the color stability of the meat. The effect of the line on growth traits was tested by using a one-way analysis of variance (General Linear Models procedure of SAS; SAS Institute, 1989). An analysis of variance was also performed to test the effect of the line, the duration of storage, and the interaction between these factors on meat quality traits. Means were compared within each factor with a significant effect, using a Newman-Keuls test for multiple mean comparisons. Pearson correlations were calculated to describe the relations between pHu, color measures, and DL of the meat.

Estimation of Genetic Parameters Descriptive statistics including the test of the normality of the distribution of meat quality parameters were obtained by the UNIVARIATE procedure of SAS (SAS Institute, 1989). Two analyses were performed to estimate the heritabilities and genetic correlations between pHu, L*, a*, b*, BW, and the absolute or the relative weights of breast meat and abdominal fat, respectively. The model included the fixed effects of hatch and sex, the direct genetic effect of each animal, and, except for meat quality parameters, the environmental maternal effect as described by Le Bihan-Duval et al. (1998). Maternal effects were not taken into account for the estimation of the genetic parameters of the meat quality traits, as it had been shown in a preliminary analysis that they resulted in a poor rate of convergence on the genetic parameters estimates. In order to take into account the selection background and its effect on the variability of BW and body composition, all the data available for these traits from the beginning of the experiment (i.e., a total of 12,090 birds) were included in the genetic analysis. Restricted maximum likelihood estimates of the genetic parameters were computed using the VCE 3.2 package of Groeneveld (1993).

RESULTS

Effect of Selection on Meat Parameters Means and standard deviations of BW and of body composition traits obtained by dissection of the selected and control birds are reported in Table 1. The data clearly show that the animals from the Quality line exhibited a higher BW and BRY (18 and 9%, respectively) and a lower AFP (–20%) than the control line. Ultimate pH differed slightly but significantly (P < 0.05) between the two lines,

824

BIHAN-DUVAL ET AL.

TABLE 1. Means and standard deviations of body weight and body composition traits in the samples used for comparing the control and selected lines Selected birds Control birds n = 66 n = 65 Line effect

Trait

Body weight, g 1,933.0 Breast meat weight, g 274.1 Breast meat yield, % 14.2 Abdominal fat weight, g 36.4 Abdominal fat, % 1.9

± 95.4 1,632.2 ± 71.7 ** ± 24.7 ± 0.9 ± 11.5 ± 0.6

213.0 ± 16.1 ** 13.0 ± 0.8 ** 38.7 ± 9.2 2.4 ± 0.5

NS **

**P < 0.001.

with average values of 5.78 ± 0.10 and 5.68 ± 0.12 for the selected and control birds, respectively. The results of the variance analysis with the effects of the line, the postslaughter time, and the interaction between these factors on color and DL of the meat are reported in Table 2. No effect of the line was found for L*. In contrast, a significant difference between both lines was observed for a* and b*, with lower values in the Quality birds. Drip losses appeared to be significantly (P < 0.05) lower in the Quality line than in the control line. Results of the analysis of variance indicated significant variations of color during storage. L* remained constant until 6 d postslaughter and then significantly increased. a* and b* significantly increased until 3 d postslaughter and then remained at a steady level. As expected, DL increased with the duration of storage, with a more pronounced difference between lines after 10 d of storage than after 3 d. A significant negative correlation was found between pHu and lightness at 24 h postmortem (–0.58, P < 0.001 in the control line; –0.60, P < 0.001 in the Quality line). Ultimate pH was also negatively correlated with DL, with correlations with DL at 3 d postslaughter of –0.41 (P < 0.001) and –0.38 (P < 0.01) in the control and Quality birds, respectively. In contrast, L* was positively correlated with

DL, with correlations of 0.57 (P < 0.001) and 0.53 (P < 0.001) in the control and Quality birds, respectively.

Genetic Variability of Meat Parameters and Genetic Correlations with Growth and Body Composition The basic statistics of the measurements of meat quality in this analysis are reported in Table 3. Skewness and kurtosis did not largely deviate from the values (i.e., zero) of a normal distribution. Additionally, results of the Shapiro-Wilk test indicated that the assumption of a normal distribution could not be rejected for these traits. Estimates of the heritability of pHu and color characteristics measured 24 h postmortem and of their genetic correlations with BW, breast meat weight or yield and abdominal fat weight or percentage are presented in Table 4. The heritability of pHu was fairly high, although the standard error of this estimate was large. Our results suggest a predominant role of genetics in the control of color, as heritability estimates of these measurements were between 0.64 and 0.81. L* was genetically negatively correlated with pHu, with a fairly high estimate of –0.65. The characteristics a* and b* were closely related, with an estimated genetic correlation of 0.72. On the other hand, these color characteristics were not correlated with L* and pHu measurements, with the exception of a moderate negative correlation (–0.45) between L* and a*. Ultimate pH was not correlated with BW and muscle development, whereas a significant negative genetic correlation was observed with abdominal fatness. Except for BRY, L* was significantly positively correlated with all the performance traits with estimated correlations ranging from 0.37 to 0.51. There were moderate but significant negative genetic relationships between a* and BW, BRW, and BRY. In contrast, abdominal fatness was found to be moderately positively correlated with a*. The same (but

TABLE 2. Effect of the line and postmortem time on color measures and drip loss of the meat. Results are expressed as x ± SEM (n = 65 and 66 for control and quality birds, respectively) Analysis of variance Postmortem time Trait

Line

1 d

3 d

6 d

10 d

Lightness L*

Control

51.16 ± 0.28

50.73 ± 0.26

51.12 ± 0.24

52.71 ± 0.24

Redness a*

Quality Control

50.83 ± 0.24 1.29 ± 0.07

51.04 ± 0.20 2.47 ± 0.06

51.25 ± 0.24 2.47 ± 0.06

52.67 ± 0.25 2.45 ± 0.06

Quality Yellowness b* Control

0.46 ± 0.08 13.50 ± 0.16

1.49 ± 0.09 14.29 ± 0.16

1.78 ± 0.09 14.70 ± 0.15

1.81 ± 0.08 14.35 ± 0.14

Quality Drip loss, % Control

12.53 ± 0.17 . . .

13.45 ± 0.18 2.04 ± 0.10

13.84 ± 0.16 3.56 ± 0.14

13.74 ± 0.16 5.19 ± 0.17

1.07 ± 0.05

2.15 ± 0.09

3.43 ± 0.13

Quality *P < 0.05. **P < 0.001.

. . .

Line effect

Time effect

Line × time interaction

NS

**

NS

**

**

NS

**

**

NS

**

**

*

825

SELECTION OF BROILERS AND MEAT QUALITY TABLE 3. Basic statistics of meat quality parameters used for the estimation of the genetic parameters

Trait

Mean (n = 231)

SD

Skewness1

Kurtosis1

pHu lightness L* redness a* yellowness b*

5.77 50.7 0.51 12.5

0.10 2.00 0.67 1.30

–0.019 –0.062 0.229 –0.137

0.296 –0.198 0.166 0.280

1Calculated by UNIVARIATE procedure of SAS (SAS Institute, 1989). Standard values of skewness and kurtosis are equal to zero for a normal distribution.

less marked) trends were observed for b*, except for a nonsignificant genetic correlation with BW.

DISCUSSION In this study, meat quality parameters are compared between selected birds and their controls and estimates of their genetic variability are provided. Water holding capacity was dramatically improved by selection for increased BRY, at least in these experimental lines. Indeed, DL, which averaged 5.2% in the control line at 10 d postslaughter, was reduced to 3.5% in the selected line. A significant relationship has previously been observed between pHu and DL in turkeys and broiler chickens (Barbut, 1997a,b). The present study also showed that meat with a lower pHu had a lower water holding capacity. However, it is unlikely that a difference in pHu entirely explains the difference in DL, because the magnitude of difference for pHu was quite small. Further investigation will be required to define the relationship between DL and the rate of decrease in pH postmortem. Considerable variability in this trait has been observed in turkeys (Sante´ et al., 1991), with a very rapid decline in pH in animals of a high performance line. McKee and Sams (1998) confirmed in turkeys that, as with pale, soft, exudative meat in the pig, rapid development of rigor mortis at a high postmortem temperature results in pale and exudative meat. Similar studies have not yet been conducted in the chicken. The significant correlation between L* of the meat and its

pHu reported previously (Barbut, 1997a,b) was also found in the present study; however, no difference in L* was observed between the control and Quality lines in the present study. The breast meat of the control and quality birds differed significantly in a* and b*, with lower values in the selected line. This result is in agreement with observations in turkey breast muscle (Sante´ et al., 1991) showing significantly decreased a* (at 1, 2, and 7 d postmortem) in a high performance line compared to a low performance line. Redness of chicken breast meat has been shown to correlate positively with myoglobin content (Boulianne and King, 1995), which is lower in glycolytic than in oxidative myofibers (Lieber, 1992). The present study also showed similar color instability with storage in both lines. A significant increase in L* was observed at 6 and 10 d postslaughter, which could result from myosin degradation, as already observed in beef meat (Hector et al. 1992). a* and b* increased in both lines during the first days of storage, in agreement with data in turkeys showing that b* increased during storage at 4 C until 7 d postslaughter (Sante´ et al., 1991). This variation may be related to the chemical changes in muscle myoglobin pigment, which is predominantly converted into purple reduced myoglobin and brown metmyoglobin during the first days postslaughter [Millar et al. (1994) in chickens, Sante´ et al. (1991) in turkeys]. This work suggests a substantial role of genetics in the control of meat quality traits, especially when color values are considered. A large number of heritability estimates have been reported for the pig and, as indicated in a review of Sellier and Monin (1994), heritabilities appear to be moderate for pHu and color. According to a recent study by Larzul (1997) in the pig, heritabilities of L* and a* characteristics appear to be higher for glycolytic-white or intermediate muscles (with heritability between 0.15 and 0.43) than for oxidative-red muscle (with almost null heritability). This result is consistent with our high estimates of heritability because the P. major muscle of chickens is an almost completely white fast-twitch muscle (Remignon et al., 1996). Lightness appeared to be highly genetically

TABLE 4. Heritabilities and genetic correlations and their approximate standard errors of meat quality parameters measured 24 h postmortem: [ultimate pH (pHu), lightness (L*), redness (a*), yellowness (b*)] and of performance traits: breast meat weight (BRW) and yield (BRY), abdominal fat weight (AFW), and percentage (AFP)

pHu L* a* b* BW BRW AFW BRY AFP

pHu

L*

a*

b*

BW

0.49* ± 0.11

–0.65* ± 0.10 0.11 ± 0.08 –0.11 ± 0.11 0.08 ± 0.75* ± 0.08 –0.45* ± 0.05 0.06 ± 0.07 0.51* ± 0.81* ± 0.04 0.72* ± 0.06 –0.25* ± 0.64* ± 0.06 0.11 ± 0.41* ±

BRW

AFW

BRY

0.07 –0.02 ± 0.06 –0.64* ± 0.10 –0.12* ± 0.08 0.37* ± 0.05 0.48* ± 0.16 0.07 ± 0.05 –0.48* ± 0.03 0.13* ± 0.05 –0.51* ± 0.06 –0.18* ± 0.06 0.36* ± 0.11 –0.44* ± 0.02 0.77* ± 0.02 0.46* ± 0.04 0.16* ± 0.51* ± 0.03 0.26* ± 0.04 . . . 0.52* ± 0.04 . . . 0.63* ±

*Significantly different from zero, when assuming a normal distribution of the estimate.

AFP 0.04 –0.76* ± 0.08 0.07 0.41* ± 0.15 0.04 0.23* ± 0.04 0.09 0.38* ± 0.11 0.05 0.19* ± 0.05 . . . . . . 0.04 –0.20* ± 0.05 0.57* ± 0.03

826

BIHAN-DUVAL ET AL.

correlated with pHu, which suggests that measurement of meat color can be valuable for selection. Lightness appears to be highly heritable and, according to these results, selecting for lower L* values could lead to higher pHu and, in turn, to better water holding capacity. This conclusion underlines the need to define the genetic parameters of color measurements and their relationship with pHu and with the rate of pH decline by analysis of larger data sets. According to our estimates of genetic parameters, pHu was nearly independent of BW and muscle development, whereas a significant negative genetic correlation was obtained with abdominal fat. This result was consistent with the slightly higher pHu observed in the Quality birds selected for decreased abdominal fat. Selection for BW, and, to a lesser extent, for increased breast muscle development, should result in increased meat lightness, whereas reducing abdominal fatness should provide the opposite result. These results could explain why no difference in L* was observed between the control and Quality lines, as the criteria applied in the selected line would be expected to have opposite effects. Finally, estimated genetic correlations between a* or b* and growth or conformation traits indicated that selection for higher BW and breast muscle development and for lower abdominal fat should result in paler meat (by decreasing a* and b*). This finding was consistent with our finding that a* and b* were reduced in the Quality birds compared to the control birds.

ACKNOWLEDGMENTS We thank A. Boucard, M. Mangin, G. Marche´, and N. Wacrenier for their valuable technical assistance. We gratefully acknowledge C. Beaumont for the discussions and helpful suggestions for the manuscript.

REFERENCES Barbut, S., 1997a. Occurrence of pale soft exudative meat in mature turkey hens. Br. Poult. Sci. 38:74–77. Barbut, S., 1997b. Problem of pale soft exudative meat in broiler chickens. Br. Poult. Sci. 38:355–358. Barton, N. F., 1994. Breeding meat type poultry for the future targets for selection, limits to performance and market requirements for chicken. Page 33–38 in: Proceedings of the 9th European Poultry Conference, 7–12 August, Glasgow, U.K. Boulianne, M., and A. J. King, 1995. Biochemical and color characteristics of skinless boneless pale chicken breast. Poultry Sci. 74:1693–1698. Chambers, J. R., 1990. Quantitative genetics and selection. Genetics of growth and meat production in chickens. Page

599–643 in: Poultry Breeding and Genetics. R. D. Crawford, ed. Elsevier, Amsterdam, The Netherlands. Groeneveld, E., 1993. REML VCE: a multivariate multimodel restricted maximum likelihood (co)variance component estimation package. Pages 83–102 in: Proceedings EC Seminar on Application of Mixed Linear Models in the Prediction of Genetic Merit in Pigs. E. Groeneveld, ed. 27 May, Mariensee, Germany. Hector, D. A., C. Brew-Graves, N. Hasse, and D. A. Ledward, 1992. Relationship between myosin denaturation and the colour of low-voltage-electrically-stimulated beef. Meat Sci. 31:299–307. Jego, Y., B. Besbes, and J. L. Donal, 1995. Analyse de la variabilite´ ge´ne´tique et de la re´ponse a` la se´lection dans deux ligne´es commerciales de poulets de chair. Page 125–127 in: Proceedings of the leres Journe´es de la Recherche Avicole. 28–30 mars, Angers, France. Larzul, C., 1997. Variabilite´ ge´ne´tique d’une mesure in vivo du potentiel glycolytique musculaire chez le porc. Relations avec les performances, les caracte´ristiques du muscle et la qualite´ technologique des viandes. The`se de Docteur de l’Institut National Agronomique, Paris-Grignon, France. Le Bihan-Duval, E., S. Mignon-Grasteau, N. Millet, and C. Beaumont, 1998. Genetic analysis of a selection experiment on increased body weight and breast muscle weight as well as on limited abdominal fat weight. Br. Poult. Sci. 39: 346–353. Lieber, R. L., 1992. Skeletal Muscle Structure and Function. J. P. Butler, ed. Williams and Wilkins, Baltimore, MD. McKee, S. R., and A. R. Sams, 1998. Rigor mortis development at elevated temperature induces pale exudative turkey meat characteristics. Poultry Sci. 77:169–174. Millar, S., R. Wilson, B. W. Moss, and D. A. Ledward, 1994. Oxymyoglobin formation in meat and poultry. Meat Sci. 36:397–406. Remignon, H., V. Desrosier, and G. Marche, 1996. Influence of increasing breast meat yield on muscle histology and meat quality in the chicken. Reprod. Nutr. Dev. 36:523–530. Ricard, F. H., G. Marche, and E. Le Bihan-Duval, 1994. Essai d’ame´lioration par se´lection de la qualite´ de carcasse du poulet de chair. INRA, Prod. Anim. 7(4):253–261. Sante´, V., G. Bielicki, M. Renerre, and A. Lacourt, 1991. Post mortem evolution in the Pectoralis superficialis muscle from two turkey breeds: relationship between pH and colour changes. Page 465–468 in: 37th International Congress of Meat Science and Technology. Vol 1. 1–6 September, Kulmbach, Germany. SAS Institute, 1989. SAS User’s Guide: Statistics. SAS Institute Inc., Cary, NC. Sellier, P., and G. Monin, 1994. Genetics of pig meat quality: a review. J. Muscle Foods 5:187–219. Vereijken, A.L.J., 1992. Genetics of body conformation and breast meat yield in broilers. Page 98–100 in: Proceedings of the XIX World’s Poultry Congress. Vol. 3. 20–24 September, Amsterdam, The Netherlands.