- Email: [email protected]
Divergent Selection for Ascites Incidence in Chickens
GENETICS Divergent Selection for Ascites Incidence in Chickens H. O. Pavlidis,*1 J. M. Balog,† L. K. Stamps,* J. D. Hughes Jr.,*2 W. E. Huff,† and N. B. Anthony*3 *Department of Poultry Science, University of Arkansas, Fayetteville 72701; and †USDA-ARS, Poultry Production and Product Safety Research Unit, Fayetteville, AR 72701 ascites appeared to be associated with livability. By generation 10, selection for ascites in line RES increased livability by 11.5 d, whereas in line SUS, livability was decreased by 8 d. Although divergent selection for ascites resulted in a reduction in d 42 BW for both the SUS and RES lines, the SUS line was approximately 163 g heavier than the RES line. Negative genetic correlations between ascites and the right ventricle:total ventricle (RV:TV) ratio were observed in both the SUS and RES lines; however, no significant change in the RV:TV ratio was observed for birds reared under normal conditions in either line. The current data raise questions about the validity of using the RV:TV ratio as an indicator trait in a selection program designed to reduce the incidence of ascites. Overall, direct selection for resistance to ascites by using sire family performance appeared to be an effective means of reducing the incidence of ascites. However, simultaneous selection for BW should be applied to counterbalance the losses in correlated BW.
Key words: chicken, ascites, divergent selection, hypobaric hypoxia, heritability 2007 Poultry Science 86:2517–2529 doi:10.3382/ps.2007-00134
INTRODUCTION Intense selection for growth, feed conversion, and white meat yield in broilers has led to significant physiological changes. One of the changes of particular importance is the increased incidence of metabolic disorders, including ascites. Ascites is an economically important disease faced by the industry over the past 2 decades. It is a 2-fold problem, influencing both the live production and processing sectors of the industry. Typically, ascites mortality can range from 0 to 30% in broiler flocks. It has been estimated that 8% of 361,000,000 broiler deaths each year can be attributed to ascites. In 2002, it was estimated that
©2007 Poultry Science Association Inc. Received March 29, 2007. Accepted July 12, 2007. 1 Current address: Nicholas Turkey Breeding Farms, 31186 Midland Trail East, Lewisburg, WV 24901. 2 Current address: Cobb-Vantress, Inc., PO Box 1030, Siloam Springs, AR 72761. 3 Corresponding author: [email protected]
these on-farm late grow-out mortalities cost more than $100 million per year in the United States. In addition, 0.05% of all processing plant condemnations can be attributed to ascites (A. Rossi, Cobb-Vantress Inc., Siloam Springs, AR, personal communication). Ascites is a metabolic disease characterized by an enlarged flaccid heart, variable liver changes, and accumulation of fluid in the abdominal cavity (Riddell, 1991). Ascites is a result of the inability of the broiler to provide its tissues with an adequate supply of oxygen (Lister, 1997). The ascites syndrome cannot be classified as a contagious or infectious disease, but rather as a progressive disease starting with pulmonary hypertension and progressing to right side congestive heart failure, with the end result of congestive heart failure (Lister, 1997; Mitchell, 1997). Several management techniques have been developed and implemented to help curb the incidence of ascites in broiler flocks. Unfortunately, these techniques are designed to slow early bird growth, thus not allowing the bird to achieve its full genetic potential. Feed restriction is one the techniques most commonly used to curb the
2517
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
ABSTRACT Chicken lines that were either resistant or susceptible to ascites syndrome were developed by using a hypobaric chamber to induce the disease. Birds were reared in a hypobaric chamber that simulated high altitude by operating under a partial vacuum, which thereby lowered the partial pressure of oxygen. Ascites mortality data from birds reared under hypobaric chamber conditions were used to select siblings to be used for breeding. The response to selection for the susceptible (SUS) and resistant (RES) lines of chickens was very rapid from the base population, which exhibited an incidence of ascites of 75.3%. Extremes in the incidence of ascites were observed in generation 8, with line SUS exhibited an average incidence of ascites of 95.1%, and in generation 9, with line RES exhibited an average incidence of ascites of 7.1%. The incidence of ascites in the relaxed line remained relatively stable and currently has a general incidence of ascites of 60%. The heritability estimates ± SE for ascites were estimated to be 0.30 ± 0.05 and 0.55 ± 0.05 for lines SUS and RES, respectively. Changes in the incidence of
2518
PAVLIDIS ET AL.
ascites and lived to sexual maturity were chosen to be the first generation in an ascites-resistant line. Successive generations were reared under cold stress conditions to screen for ascites susceptibility. After 2 generations, progeny from this RES line had significantly lower ascites mortality compared with the base population. The males exhibited an ascites mortality of 6.4% and the females 0% compared with the base populations, which exhibited an ascites mortality of 43.6 and 12.3% respectively. This unidirectional selection study shows that ascites resistance can be successfully selected against. However, in the study by Wideman and French (2000), valuable information such as heritabilities and genetic correlations were not estimated and correlated responses were not measured. Divergent selection for ascites has been practiced in the authors’ laboratory through the use of a hypobaric chamber, which simulates high-altitude conditions, to repeatably induce ascites, thereby allowing exploration of the genetic relationship and correlated responses associated with selection for ascites. The purpose of the current research was to summarize the results of 10 generations of selection for ascites.
MATERIALS AND METHODS Hypobaric Chamber The hypobaric and local altitude chambers were each 2.4 × 3.7 × 2.4 m, and each was capable of housing 240 broilers to 6 wk of age. The chambers were equipped with identical custom stainless steel batteries fitted with trough feeders and nipple waterers. The chambers were matched in terms of ventilation rate, temperature, and size. The environmental variables monitored included altitude, ventilation, temperature, and humidity. Ventilation was set to maintain a constant airflow at the rate of 17 m3/min. Birds were warm room brooded, with temperatures decreased weekly. The hypobaric chamber was set to simulate 2,900 m (9,500 ft) above sea level. A control environment was maintained in a local altitude chamber at 390 m (1,200 ft) above sea level. All birds placed at simulated high altitude remained at high altitude for the duration of the experiments. Daily management tasks and weighings were conducted in the respective rearing environment.
Initial Line Formation Figure 1 illustrates the development of the ascites-susceptible (SUS) and ascites-resistant (RES) lines from a commercial pedigree elite line (REL). In 1995, 3 separate hatches of pedigree male line chicks were obtained from a primary poultry breeding company. The chicks obtained in these hatches represented offspring from 16 sire families in the pedigree male line. This pedigree male line had undergone one generation of relaxed selection prior to becoming the base population for development of the ascites lines. The chicks obtained in the first and third hatches were reared under typical broiler management conditions in the
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
incidence of ascites (Julian, 1998; Balog et al., 2000; Roush and Wideman, 2000; Wideman, 2000; Balog, 2003). Another technique used is intermittent lighting (Julian, 2000). Ascites had been a difficult syndrome to study because of the problems associated with obtaining a uniform characteristic flock incidence. Thus, to better understand both the physiological and genetic components, a repeatable model needed to be developed. Both surgical and nonsurgical methods have been developed. Surgical methods include clamping of the left pulmonary artery (Wideman and Kirby, 1995) and unilateral occlusion of the primary bronchus (Wideman et al., 1997). Wideman and Erf (2002) reported that the intravenous injection of microparticles can also be an effective method of inducing ascites. Nonsurgical methods of inducing ascites include various cold stress models, which involve exposing the bird to constant cool temperatures (Julian et al., 1989; Vanhooser et al., 1995), to a gradual decrease in temperature (Verstegen et al., 1989; Buys et al., 1999), or to an episodic cold stress (Shlosberg et al., 1996). Low-ventilation models, which simulate a winterized broiler house, have also been implemented, but this method does not induce ascites to the same magnitude as cold stress (Julian and Wilson, 1992; Shlosberg et al., 1992). High-altitude simulation through the use of a hypobaric chamber is another nonsurgical model producing a consistent incidence of ascites in broilers (Owen et al., 1990; Witzel et al., 1990; Balog et al., 2000; Anthony et al., 2001). By operating under a partial vacuum, a reduction of the partial pressure of oxygen is created, thus simulating conditions observed in naturally high altitudes. Foundation experiments involving the hypobaric chamber allowed only a small number of individuals to be tested (Owen et al., 1990). The use of larger hypobaric chambers with tight environmental conditions has increased this model’s usefulness (Balog et al., 2000; Anthony et al., 2001). In addition, constant monitoring of environmental variables such as altitude, ventilation, temperature, and humidity provides the researcher more environmental control. This allows the researcher the ability to repeatedly “dial in” the same environment every time the chamber is used, thus making it one of the most ideal noninvasive models for inducing ascites in broilers. Ascites is known to be influenced by both environmental (Julian, 2000) and genetic factors (Lubritz et al., 1995; Wideman and French, 2000; Anthony et al., 2001) and is an artifact of intense genetic selection for rapid growth rate, feed conversion, heavy BW, and white meat yield in broiler populations. Lubritz et al. (1995) used cold stress as the model to induce ascites in 3 commercial male lines and reported moderate to high heritabilities (0.11 to 0.44) for ascites score, and moderate heritabilities (0.21 to 0.27) for the ratio of the right ventricle (RV) to total heart weight. Therefore, it seemed likely that a selection program could be developed and be successfully implemented to reduce the incidence of ascites in broiler populations. Wideman and French (2000) used the pulmonary artery clamp to screen a small population of broilers for ascites susceptibility. The individuals that did not succumb to
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
2519
hypobaric and local altitude chambers. Birds reared in the hypobaric chamber were used to generate ascites mortality data, which was used to make sib selection decisions. Birds reared in the local altitude chamber served as a control for the altitude challenge. All birds were provided ad libitum consumption of water and a corn and soybean-based broiler starter and grower ration that was formulated to meet or exceed NRC (1994) requirements. The photoperiod consisted of 23 h of light and 1 h of dark. All the hypobaric chamber daily management tasks and weighings were conducted under partial vacuum through the use of an air lock, which allowed for pressure equilibrium. Growth was evaluated by measuring BW at d 21 and 42 (BW21 and BW42, respectively) on all survivors. Birds were checked daily and all mortalities were necropsied to determine the cause of death. At necropsy, mortalities were classified as ascitic or nonascitic. Mortalities were evaluated for symptoms of ascites, such an enlarged and flaccid heart, liver lesions, presence of fluid in the pericardium, and presence of fluid in the abdominal cavity. Mortalities were classified as ascitic if fluid was present in the abdominal cavity. The BW at death, sex, and day of death in the study were also recorded at the time of necropsy. At 6 wk, all remaining survivors in both the local and hypobaric chambers were euthanized by cervical dislocation and necropsied to check for the presence of ascites. Individual hearts, livers, and spleens were removed and individual organ weights recorded. The hearts were dissected into the RV and the remaining heart, and RV were
individually weighed to calculate the RV to total ventricle (TV) ratio (RV:TV ratio), which has been shown to be an indicator of ascites (Julian, 1993; Wideman and Bottje, 1993; Lubritz et al., 1995). Birds were considered ascitic if fluid was present in the abdominal cavity. Sire family selection was applied to develop the SUS and RES lines. Average ascites mortality was calculated for each of the original 16 sire families. The 8 sire families with the lowest average ascites mortality percentages were selected to become the RES line, and the 8 sire families with the highest average ascites mortality percentages became the SUS line. The REL line was maintained as a control and is currently maintained as a randomly mated population with the stipulation that full- and half-sib matings are not permitted. At the same time that chicks from the first and third hatches were being challenged in the hypobaric chamber, chicks obtained from the second hatch were placed in floor pens and grown as pedigreed broilers. These chicks were grown under typical broiler breeder management conditions, as suggested by the primary breeder from which they originated. Up to 5 wk of age, birds were allowed ad libitum access to a corn and soybean-based broiler ration that was formulated to meet or exceed NRC (1994) requirements. At 5 wk, individual BW were obtained; birds were sex separated and placed on a feed and photoperiod restriction program to delay the onset of lay. Water was available for ad libitum consumption throughout the entire rearing period.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 1. Development of the ascites-susceptible and ascites-resistant lines for a commercial pedigree elite line based on the performance of progeny when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level.
2520
PAVLIDIS ET AL.
At 20 wk of age, average ascites mortality data obtained in the altitude challenges were used to randomly select 3 males from each of the 8 sire families from the SUS and RES lines. This resulted in the creation of 24 sire families within each of the respective lines. Each selected male was housed in individual broiler breeder male cages and randomly assigned to 3 randomly chosen females from the same line, with the stipulation that full-sib and halfsib matings would not be permitted. Matings were performed by artificial insemination, and complete pedigree information was collected. The respective lines were closed after their initial formation.
Current Selection Methods The method of selection and propagation of the ascites lines is illustrated in Figure 2. All matings were performed by artificial insemination, and complete pedigree information was collected. Each generation, 2 hatches of offspring were challenged at high altitude in the hypobaric chamber. Hypobaric mortality information was used to select sibs grown under normal industry conditions. These pedigree replacements were derived from 2 to 4 pedigree hatches from the same sire families that provided the hypobarically challenged chicks. The floor-reared pedigree replacements were maintained under typical broiler breeder management condi-
tions as described during initial line formation. Briefly, birds were allowed ad libitum access to an NRC (1994)formulated broiler ration up to 5 wk of age. At 5 wk, individual BW were obtained, and birds were separated by sex and placed on a feed and photoperiod restriction program to control growth and promote optimal reproductive performance later in life. Water was available for ad libitum consumption throughout the entire rearing process. For the altitude challenge, birds were reared under typical broiler management conditions and allowed ad libitum access to water and an NRC (1994)-formulated corn and soybean-based starter and grower ration. The photoperiod consisted of 23 h of light and 1 h of dark. All the hypobaric chamber daily management tasks and weighings were conducted under partial vacuum through the use of an air lock, which allowed for pressure equilibrium. Growth was evaluated by measuring the BW of all survivors at 3 and 6 wk. At 6 wk, all remaining survivors were euthanized by cervical dislocation and necropsied to check for the presence of ascites as described previously. From the altitude challenge, average ascites mortality for each sire family was calculated and used to select breeders from the floor-reared sibs. For each line, breeder males were randomly selected from no more than the top 6 sire families, and females from no more than the top 10 sire families. The selected breeders were placed in individual
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 2. Selection schematic for the ascites-susceptible and ascites-resistant lines by using hypobaric exposure in a hypobaric chamber set at 2,900 m above sea level as the method of inducing ascites.
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
2521
cages between 18 and 20 wk of age. A sire family for each line was constructed by randomly assigning 3 females to each of 24 males with the stipulation that full-sib and halfsib matings were not permitted.
Data Analysis The change in traits over generations for lines SUS, RES, and REL were estimated by linear regression of line means on generations by using the procedure regression feature in SAS software (SAS Institute, 1988). Data were analyzed according to the multitrait derivative free restricted maximum likelihood procedures outlined by Boldman et al. (1993). An animal model was used in both lines SUS and RES to estimate heritabilities and genetic correlations. For ascites, RV:TV ratio, and livability, no fixed effects or covariates were used. For BW21 and BW42, ascites was used as a covariate and sex was used as a fixed effect. Generation was not used in any model to estimate heritabilities or genetic correlations in either line (L. D. Van Vleck, Animal Science Department, University of Nebraska, Lincoln, personal communication).
RESULTS AND DISCUSSION Ascites Mortality The changes per generation for the incidence of ascites in lines SUS, RES, and REL are presented in Figure 3. Significant linear regression coefficients were observed in all 3 lines for the percentage of ascites mortality. For line SUS, there was a significant average increase in ascites mortality of 0.95% per generation, whereas in line RES
there was a significant decrease in ascites mortality of 6.75% per generation. Although no selection pressure was applied to line REL for any traits, a significant reduction in ascites mortality of 3.14% per generation was observed. The response to selection for the ascites SUS and RES lines of chickens was very rapid, with divergence for ascites mortality beginning to occur in generation 3. To date, extremes in the incidence of ascites have been observed in generation 8 for line SUS (95.14%) and in generation 9 for line RES (7.14%). The incidence of ascites in line REL has decreased slightly; however, it has had an overall average incidence of approximately 60%. The response to selection for incidence of ascites in lines SUS and RES was extremely rapid, suggesting that a large genetic component may be associated with ascites mortality. Therefore, the heritability for ascites was expected to be high. Heritability estimates for ascites mortality in lines SUS and RES are presented in Tables 1 and 2, respectively, and as expected, were found to be high. Lubritz et al. (1995) were the first to report heritabilities for ascites in birds reared under cold stress conditions, and these heritabilities ranged from 0.11 to 0.44, depending on the genetic line tested. Other studies (De Greef et al., 2001; Pakdel et al., 2002) in which birds were reared under cold stress have reported heritability estimates consistent with those reported by Lubritz et al. (1995). The heritability estimates reported in the current study are consistent with those reported for birds reared under cold stress conditions (Lubritz et al., 1995; De Greef et al., 2001; Pakdel et al., 2002). The heritability for ascites mortality was higher in line RES than in line SUS (Tables 1 and 2). These differences in heritabilities can be attributed to the differences in response to selection for ascites mortality. The response to
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 3. Incidence of ascites mortality per generation for lines susceptible (SUS) and resistant (RES) to ascites and for a nonselected control line (REL) when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level.
2522
PAVLIDIS ET AL. Table 1. Estimates of heritabilities, genotypic correlations, and phenotypic correlations of traits measured under hypobaric conditions in the ascites-susceptible line of chickens1 Trait Ascites RV:TV Livability BW21 BW42
Ascites
RV:TV
Livability
BW21
BW42
0.30 (0.05) NC2 0.73 0.28 0.18
−0.27 0.35 (0.10) 0.00 NC −0.23
0.87 −0.03 0.47 (0.05) 0.64 −0.06
0.55 NC 0.58 0.66 (0.07) 0.71
0.71 0.19 0.43 0.71 0.55 (0.12)
1 Genetic correlations are above the diagonal, phenotypic correlations are below the diagonal, and heritabilities (SE) are on the diagonal. RV:TV = right ventricle:total ventricle weight; BW21 = BW at d 21; BW42 = BW at d 42. 2 NC = no convergence obtained.
Livability The change per generation for livability in lines SUS, RES, and REL are presented in Figure 4. Livability is defined as how long a bird is able to live in the hypobaric chamber during a 42-d period without succumbing to ascites. Significant linear regression coefficients were observed for all lines for livability. Over 10 generations, selection in line RES has extended livability by 11.5 d, whereas in line SUS, livability has been decreased by 8 d. These changes in livability in lines RES and SUS appear to be associated with the timing of the onset of ascites for the respective lines. Figure 5 clearly shows that selection for ascites has shifted the ascites mortality curves, resulting in line SUS succumbing to ascites earlier than line RES. Line SUS typically has ascites mortalities beginning at approximately d 5, whereas line RES does not have ascites mortalities until approximately d 11. On average, birds in
line REL had an average day of death from ascites at 27 d, whereas lines SUS and RES were at 19 and 22 d, respectively. Therefore, selection in both lines has resulted in birds succumbing to ascites earlier than those in line REL. This is especially beneficial in line RES, because ascites mortalities that occur are relatively early in the growout period, when resources such as feed and economic loss are minimal. However in line REL, the average livability was 29 d, whereas livability in line SUS was decreased to 22 d and was increased in line RES to 39 d. These changes in livability are directly related to the respective ascites mortalities observed for lines SUS and RES (Figure 3). The response in livability was extremely rapid, and livability appears to be correlated with ascites mortality in both lines SUS and RES. Therefore, the heritability for livability was expected to be high, and positive genetic and phenotypic correlations were expected between ascites mortality and livability. The heritability estimates for livability as well as the genetic and phenotypic correlations between ascites mortality and livability are presented in Tables 1 and 2. Heritabilities were found to be high, and there were strong, positive genetic and phenotypic correlations between ascites incidence and livability in lines SUS and RES. Therefore, selection for ascites susceptibility will result in a reduction in livability, whereas selection for ascites resistance will increase livability.
BW at Hatch, d 21, and d 42 The relationship between ascites and BW has been well documented (Julian et al., 1986; Julian, 2000; Balog, 2003). Birds that have heavier BW and exhibit faster growth rates
Table 2. Estimates of heritabilities, genotypic correlations, and phenotypic correlations of traits measured under hypobaric conditions in the ascites-resistant line of chickens1 Trait Ascites RV:TV Livability BW21 BW42
Ascites
RV:TV
Livability
BW 21
BW 42
0.55 (0.05) 0.25 0.81 0.24 −0.05
−0.047 0.59 (0.06) 0.52 NC −0.08
0.92 0.19 0.70 (0.5) 0.46 −0.07
0.20 NC2 0.37 0.89 (0.05) 0.62
−0.23 0.13 −0.25 0.72 0.56 (0.07)
1 Genetic correlations are above the diagonal, phenotypic correlations are below the diagonal, and heritabilities (SE) are on the diagonal. RV:TV = right ventricle:total ventricle weight; BW21 = BW at d21; BW42 = BW at d 42. 2 NC = no convergence obtained.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
selection for ascites mortality was greater in line RES than in line SUS. This difference in the response for ascites mortality may be because line RES has a greater range to reduce ascites mortality than the range line SUS has to increase ascites mortality, because ascites mortality was 75.3% in generation 0. Obviously, ascites mortality cannot exceed greater than 100% for line SUS or less than 0% for line RES. With an average incidence of ascites of 95.1%, line SUS began to reach a physiological ceiling by generation 8 (Figure 3). Approaching this physiological ceiling limits the amount of progress that can be made in this line; thus, heritability for the trait would begin to decrease, consistent with what was seen with the heritability estimates for line SUS as compared with line RES (Tables 1 and 2).
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
2523
are more likely to develop ascites than those birds with lighter BW and slower growth rates (Julian et al., 1986; Julian, 2000; Balog, 2003). Because there is a defined negative relationship between ascites and BW, it is important to understand how selection for ascites susceptibility and resistance influences BW, a trait of high economic importance to the industry. The effects of selection on hatch weight, BW21, and BW42 in both the hypobaric chamber and at sea level are presented in the various panels of Figure 6. For hatch weight (Figure 6A), significant negative linear regression coefficients were observed in lines SUS and RES; however,
no significant change was observed for line REL. This resulted in line RES producing chicks that were slightly heavier than those in line SUS at hatch. However, when the initial 16 sire families were divided into lines SUS and RES in generation 0, it was unclear whether selection for susceptibility to ascites had resulted directly in a reduction in hatch weight in line SUS or whether these differences in hatch weight could be attributed to a potential founder effect (Falconer and Mackay, 1996). Another possible explanation for the differences seen in hatch weight could be the impact of selection on incubation time. Christensen et al. (2000) reported the effects of selection in various lines
Figure 5. Typical ascites mortality curve for lines susceptible (SUS) and resistant (RES) to ascites and for a nonselected control line (REL) when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 4. Livability per generation for lines susceptible (SUS) and resistant (RES) to ascites and for a nonselected control line (REL) when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level.
2524
PAVLIDIS ET AL.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 6. Change per generation for hatch weight, BW at 21 d (BW21), and BW at 42 d (BW42) for lines susceptible (SUS) and resistant (RES) to ascites and for a nonselected control line (REL) reared under normal environmental conditions in the sea level chamber and when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level. A) Hatch weight per generation for ascites lines SUS and RES and for the nonselected control line REL under normal environmental conditions. B) BW21 per generation for ascites lines SUS and RES and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. C) BW42 per generation for ascites lines SUS and RES and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. D) BW21 per generation for ascites lines SUS and RES and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber. E) BW42 per generation for ascites lines SUS and RES and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber.
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
died from ascites, whereas a smaller number had died from ascites in line RES. Thus, those birds in line SUS that had survived the 42-d challenge in the hypobaric chamber could be considered the least susceptible birds in line SUS. For those birds in line SUS to survive the hypobaric challenge, they would have needed growth rates and final BW that were similar to those of line RES, although after 10 generations, the survivors in line SUS had BW that were approximately 32 g heavier than the survivors of line RES. The impact on BW42 in line REL is approximately half that seen in lines SUS and RES. Because the incidence of ascites seen in line REL was intermediate to those seen in lines SUS and RES, the impact on BW42 for line REL in the hypobaric chamber would be expected to be intermediate to the effects seen for lines SUS and RES. For BW42 at sea level (Figure 6C), significant negative linear regression coefficients were observed across all lines. Line SUS did maintain a superior BW across both lines RES and REL. This superiority in BW for line SUS was supported by both the high, positive genetic and phenotypic correlations between ascites and BW42 (Table 1). By generation 10, line SUS had lost approximately 214 g in BW42, whereas line RES had lost 377 g. This resulted in line SUS having BW that were approximately 163 g heavier than those of line RES after 10 generations of selection. This reduction in BW in line RES was consistent with the negative genetic and phenotypic correlations reported between ascites and BW42 for line RES (Table 2). The superiority in BW for line SUS combined with the positive genetic and phenotypic correlations were consistent with the relationships that have been reported between ascites susceptibility and BW (Julian et al., 1986; Lubritz et al., 1995; Julian, 2000; Pakdel et al., 2002; Balog, 2003). Birds with heavier BW at d 21 typically had higher RV:TV ratios and were more likely to develop ascites than those birds that had lighter BW at the same age (Julian et al., 1986). Under normal environmental conditions, the increase in BW21 observed for line SUS and the decrease seen in line RES were consistent with the findings of Julian et al. (1986). Although significant negative linear regression coefficients were observed for all lines for BW42 under normal conditions, line SUS did maintain superiority for BW42 over both lines REL and RES (Figure 6C). When BW42 in line SUS was expressed as a deviation from the base population line REL, positive BW gains were detected (data not shown). Beker et al. (2003) reported unpublished data from Teeter and Wiernusz that stated the oxygen requirements for 1 g of lean and of fat accretion were 3.9 and 1.2 L, respectively, to 42 d of age in commercial broilers from 1994. We hypothesized that, in our laboratory, the differences in BW21 and BW42 seen in the current study between lines SUS and RES under normal conditions were due mainly to differences in lean tissue mass as opposed to large differences in fat mass. Under this hypothesis, the demand for oxygen would be increased in line SUS and decreased in line RES. Therefore, when reared under normal environmental conditions, when oxygen was not a limiting factor, growth was optimized, resulting in line
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
of turkeys and their respective randombred control lines, and found that selection had modified incubation time. From those data, one could hypothesize that line RES had delayed hatching, thus resulting in wetter chicks that would be heavier than earlier hatching chicks from line SUS. Although this may be a potential explanation for the differences in hatch weight between lines SUS and RES, further research is needed to determine why this difference is present. In the hypobaric chamber for BW21 (Figure 6D), significant negative linear regression coefficients were observed for lines SUS and RES, with line SUS having a larger reduction in BW per generation than line RES. No significant change was observed in BW21 for line REL. By generation 10 in line RES, there was a reduction in BW of approximately 47 g; this reduction in BW mimics current management strategies designed to minimize the incidence of ascites. These management strategies are designed to slow early growth and allow for the development of a cardiopulmonary system that is capable of supporting rapid growth while minimizing the impact on final BW (Julian, 1998, 2000; Balog et al., 2000; Roush and Wideman, 2000; Balog, 2003). By generation 10, line SUS was approximately 29 g lighter than line RES. One must remember that by d 21, more than 50% of the birds in line SUS had died from ascites, and a larger portion of the remaining birds were in various stages of progression of ascites than in line RES. Because a larger portion of birds in line SUS were having to deal with ascites compared with those in line RES, more resources in line SUS were being directed toward maintenance to try to cope with ascites. This reallocation of resources toward maintenance resulted in a reduction in the amount of resources being devoted to growth (Dunnington and Siegel, 1998; Siegel, 1999) and could be a potential explanation for the higher reduction in BW seen in line SUS compared with line RES. For BW21 at sea level (Figure 6B), significant linear regression coefficients were observed for all lines. Gains in BW per generation were observed for lines SUS and REL, whereas there was a reduction in BW for line RES. By generation 10, line SUS was approximately 112 g heavier than line RES. Julian et al. (1986) reported a positive relationship between ascites and BW at d 21 and 28, suggesting that selection for increased susceptibility to ascites would result in an increase in BW21. Again, as seen with line RES in the hypobaric chamber (Figure 6D), under normal conditions in which oxygen is not a limiting factor, there was a significant reduction in BW. This reduction in early BW under normal conditions further supports the fact that selection for ascites resistance has resulted in a pattern of growth that mimics current industry practices for management of ascites. For BW42 (Figure 6E) in the hypobaric chamber, significant negative linear regression coefficients were observed across all lines. Selection had the greatest impact on BW in line RES, although line SUS had final BW42 that were similar to those of line RES. This similarity in final BW between lines RES and SUS was not unexpected, because by d 42 the majority of the birds placed in line SUS had
2525
2526
PAVLIDIS ET AL.
Heart Variables The impact of selection for resistance and susceptibility to ascites on the right and left ventricles relative to BW, as well as its impact on the RV:TV ratio in both sea level and hypobaric conditions, are presented in the various panels of Figure 7. Across all lines for birds reared at sea level, no significant changes were observed for relative RV (Figure 7B). For relative left ventricle, no significant changes were observed for lines SUS and RES; however, there was a slight reduction in line REL (Figure 7C). For those birds reared in the hypobaric chamber, significant linear regression coefficients were observed for relative RV for lines SUS and RES, whereas no significant changes were observed in line REL (Figure 7E). The reduction per generation for relative RV in line RES was 3 times greater than that observed in line SUS (Figure 7E). For relative left ventricle, significant negative linear regression coefficients were observed across all lines, and the change in relative left ventricle in line SUS was 3 times greater than that of line RES (Figure 7F).
Because the available oxygen is significantly reduced in hypobaric conditions, the natural response is to increase red blood cell production to increase the oxygen-carrying capacity (Balog et al., 2000). This increase in red blood cell production results in an increase in blood hematocrit and, in turn, causes an increase in blood viscosity (Balog et al., 2000). This increase in blood viscosity results in the heart having to work harder to pump this viscous blood to the tissues, resulting in ventricular hypertrophy, more specifically, RV hypertrophy (Julian, 2000; Wideman 2000; Balog, 2003). As the demand for oxygen increases, as seen in rapidly growing broilers, ventricular hypertrophy is greater than that of slower growing birds. This makes the faster growing birds more likely to succumb to heart failure and eventually death (Julian, 2000; Wideman, 2000; Balog, 2003). From the data in the present selection experiment, selection for resistance and susceptibility to ascites under normal conditions did not result in a significant impact on the relative right and left ventricles; however, it did result in a disproportionate change in the right and left ventricles in lines SUS and RES under hypobaric conditions. These disproportionate changes in the right and left ventricles under hypobaric conditions in lines SUS and RES are important in understanding the physiological changes that occur in the heart and how it is related to the bird’s ability to survive an ascites challenge, and thus require further investigation. At sea level, no significant changes were observed for the RV:TV ratio for lines SUS and RES (Figure 7A). Because no significant changes were observed for the relative right and left ventricles in lines SUS and RES (Figures 7B and 7C), no changes in the RV:TV ratio were expected (Figure 7A) for these lines. However, there was a slight reduction in the RV:TV ratio for line REL (Figure 7A). In the hypobaric chamber, significant negative linear regression coefficients were observed for all lines, with line RES having the greatest reduction in RV:TV ratio over 10 generations of selection (Figure 7D). The RV:TV ratio has been widely used as an indicator trait for ascites (Wideman and Bottje, 1993; Julian, 2000; Wideman, 2000). An RV:TV ratio of greater than 0.27 has been used as a cutoff point for determining the ascites status of a bird, with birds with RV:TV ratios greater than 0.27 being considered ascitic (Wideman and Bottje, 1993; Julian, 2000; Wideman, 2000). Positive genetic and phenotypic correlations between ascites and RV:TV ratio for birds reared under cold stress has been reported, and these correlations suggest that selection for decreased ascites mortality will result in a reduction in the RV:TV ratio (Lubritz et al., 1995; Pakdel et al., 2002). Although a low but positive phenotypic correlation for ascites and RV:TV ratio was observed in line RES (Table 2), under normal conditions no significant changes for the RV:TV ratio were observed in the selected lines (Figure 7A). However, line SUS had mean RV:TV ratio values that were greater than those of line RES (Figure 7A). This lack of change in RV:TV ratio for the lines was not consistent with what has been hypothesized regarding the impact of selection for ascites on the RV:TV ratio (Lubritz et al., 1995, Pakdel et al., 2002).
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
SUS having BW21 and BW42 that were greater than that of line RES. Under this hypothesis, the potential differences between lines SUS and RES in oxygen demand for growth were further exacerbated when these lines were reared in the hypobaric chamber, where oxygen was the limiting factor. However, when these lines were placed in the hypobaric chamber, where oxygen was a limiting factor, line RES, through its reduction in BW, specifically early BW, had in turn a reduced demand for oxygen and was better able to handle the challenge of the oxygen-limiting hypobaric chamber. Line SUS, which had increased BW, and in turn, increased oxygen demand to support the increase in BW observed at d 21 and 42, was unable to meet the demand because of limited oxygen availability, which led to a cascade of events, ultimately ending in death from ascites. This hypothesis regarding the impact of selection on BW and its impact on the demand for oxygen is consistent not only with the observed ascites mortality (Figure 3), but also with livability (Figure 4) and the timing of when birds succumbed to ascites in the hypobaric chamber (Figure 5). These reductions in BW21 and BW42 in line RES were consistent with current management practices, which are designed to slow early growth of the bird to allow the cardiopulmonary system to develop, often at the cost of the genetic potential of the birds. This reduction in BW for line RES has potential economic implications that could limit integrating selection for ascites resistance into a breeding program. Therefore, a geneticist must determine whether the benefit of having resistance to ascites outweighs the loss in economically important BW, and potentially other traits that may be related to ascites. For selection for ascites resistance to be economically beneficial, selection for BW must occur simultaneously. However, the addition of BW selection and any other production traits will reduce the progress that is made for resistance to ascites.
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
2527
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 7. Change per generation for the right ventricle:total ventricle (RV:TV) ratio, right ventricle (RV) relative to BW at 42 d (BW42), and left ventricle relative to BW42 for lines susceptible (SUS) and resistant (RES) to ascites and a nonselected control line (REL) reared under normal environmental conditions in the sea level chamber and when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level. A) RV:TV ratio per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. B) RV relative to BW42 per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. C) Left ventricle relative to BW42 per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. D) RV:TV ratio per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber. E) RV relative to BW42 per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber. F) Left ventricle relative to BW42 per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber.
Although no significant differences were observed for the change in the RV:TV ratio in lines SUS or RES (Figure 7A), the differences in mean RV:TV values could be a result of differences between the initial 16 sire families for the
RV:TV ratio when they were split into lines SUS and RES. However, under normal conditions, none of these mean values exceeded the 0.27 threshold once divergence for ascites mortality had occurred (Figure 7A).
2528
PAVLIDIS ET AL.
The current data, combined with the negative genetic correlations and low positive phenotypic correlations reported in the current study between ascites and RV:TV ratio for lines SUS and RES (Tables 1 and 2), suggested that selection for resistance and susceptibility to ascites has not directly modified the RV:TV ratio. Although heritability of the RV:TV ratio was high, the phenotypic correlations, combined with the negative genetic correlations between the RV:TV ratio and ascites (Tables 1 and 2), suggested that sole selection for the RV:TV ratio may not be the best means for reducing the incidence of ascites in pedigree populations.
Inbreeding
General Synthesis The response to selection for the SUS and RES lines of chickens was very rapid from the base population, which exhibited an incidence of ascites of 75.30% (Figure 3). To date, extremes in the incidence of ascites were observed in generation 8 for line SUS (95.14%) and in generation 9 for line RES (7.14%; Figure 3). The heritabilities for ascites were estimated to be 0.30 ± 0.05 and 0.55 ± 0.05 for lines SUS and RES, respectively (Tables 1 and 2). These heritability estimates are consistent with what has previously been reported for ascites measured under cold stress conditions (Lubritz et al., 1995; De Greef et al., 2001; Pakdel et al., 2002). The rapid response seen for the incidence of ascites and the moderate to high heritabilities suggest that a few major genes may control ascites. Selection for resistance and susceptibility to ascites seems to be associated with livability. By generation 10, selection for ascites in line RES had increased livability by 11.5 d, whereas in line SUS, livability had been decreased by 8 d (Figure 4). Selection for ascites susceptibility also decreased the number of days it took for birds in line SUS to succumb from ascites, whereas in line RES it increased the number of days (Figure 5). Previously reported positive genetic correlations between ascites and the RV:TV ratio suggest that selection for a decreased RV:TV ratio will result in a reduction in the incidence of ascites (Lubritz et al., 1995). However, in the current study negative genetic correlations between ascites and the RV:TV ratio were observed in both lines SUS and RES. In addition, after 10 generations there has been no significant change in the RV:TV ratio in lines SUS and RES under normal conditions. The current data raise questions about the validity of using the RV:TV ratio as the sole indicator trait in a selection program designed to reduce the incidence of ascites. Under normal conditions, selection for ascites susceptibility and resistance has resulted in modifications of both
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Inbreeding was calculated by using pedigree information from the floor-reared sibs in an animal model (Boldman et al., 1993). The average changes in inbreeding per generation were 5.01, 5.12, and 2.14%, respectively, for lines SUS, RES, and REL.
BW21 and BW42 for lines SUS and RES (Figure 6). The modification of BW21 (reduced in line RES and increased in line SUS) was consistent with the relationship between BW21, RV:TV ratio, and ascites mortality reported by Julian et al. (1986). Although significant negative linear regression coefficients were observed for all lines under normal environmental conditions for BW42, line SUS did maintain superiority in BW42 over lines RES and REL. Under normal conditions, if BW42 in line SUS were to be expressed as a deviation from the base population line REL, then positive gains for BW42 would be observed. This increase in BW42 in line SUS under normal conditions was in agreement with the previously documented relationships between ascites mortality and BW (Julian et al., 1986; Julian, 2000; Balog, 2003; Pakdel et al., 2005). The reduction in both BW21 and BW42 seen in line RES when reared under normal conditions mimics current industry practices for minimizing the incidence of ascites. Current industry practices are designed to slow early growth to allow a sufficient cardiopulmonary system to develop, often resulting in the bird being unable to maximize its genetic potential (Julian, 1998; Balog et al., 2000; Roush and Wideman, 2000; Balog, 2003). The reduction in BW21 and BW42 in line RES was clearly related to both ascites mortality and livability (Figures 3 and 4), as well as the timing of when birds succumb to ascites in the limited oxygen environment of the hypobaric chamber (Figure 5). Overall, direct selection for resistance to ascites by using sire family performance appeared to be an effective means of reducing the incidence of ascites in broilers. However, selection for resistance to ascites did result in a reduction in both BW21 and BW42 when the birds were reared under normal environmental conditions. Therefore, a geneticist must determine whether the benefits of having resistance to ascites outweigh the loss seen in BW and the potential loss in other economically important traits. Pakdel et al. (2005) reported that selection that included ascites-related traits such as hematocrit and RV:TV ratio, when measured under both normal and cold stress conditions, would result in an increase in BW but in no genetic change for ascites mortality. This result suggested that current levels of ascites mortality in breeding populations could be maintained without loss of BW (Pakdel et al., 2005). From the data reported in the current selection program, selection to reduce ascites mortality clearly came at the cost of reducing BW. Therefore, to make genetic improvement in a breeding population for ascites mortality without a loss in BW, concomitant selection for BW and other production parameters would have to be performed. One will have to determine breeding goals while realizing that the amount of selection pressure applied to these traits will directly influence the amount of progress that is made for ascites resistance and BW. In the current study, only the generational response to selection for ascites susceptibility and resistance on various heart parameters and BW measurements is reported. Further research needs to be completed to gain a better understanding of how selection for susceptibility and resistance to ascites influences other economically important produc-
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
tion traits, such as growth, feed conversion, meat yield, and quality.
REFERENCES
Lubritz, D. L., J. L. Smith, and B. N. McPherson. 1995. Heritability of ascites and the ratio of right to total ventricle weight in broiler breeder male lines. Poult. Sci. 74:1237–1241. Mitchell, M. A. 1997. Ascites syndrome: A physiological and biochemical perspective. World’s Poult. Sci. 53:61–64. NRC. 1994. Nutrient Requirements of Poultry. 9th rev. ed. Natl. Acad. Press, Washington, DC. Owen, R. L., R. F. Wideman, A. L. Hattel, and B. S. Cowen. 1990. Use of a hypobaric chamber as a model system for investigating ascites in broilers. Avian Dis. 34:754–758. Pakdel, A., P. Bijma, B. J. Ducro, and H. Bovenhuis. 2005. Selection strategies for body weight and reduced ascites susceptibility in broilers. Poult. Sci. 84:528–535. Pakdel, A., J. A. M. van Arendonk, A. L. J. Vereijken, and H. Bovenhuis. 2002. Direct and maternal genetic effects for ascites-related traits in broilers. Poult. Sci. 81:1273–1279. Riddell, C. 1991. Developmental, metabolic, and miscellaneous disorders. Pages 839–841 in Diseases of Poultry. 9th ed. B. W. Calnek, H. J. Barnes, C. W. Beard, W. M. Reid, and H. W. Yoder Jr., ed. Iowa State Univ. Press, Ames. Roush, W. B., and R. F. Wideman Jr. 2000. Evaluation of broiler growth velocity and acceleration in relation to pulmonary hypertension syndrome. Poult. Sci. 79:180–191. SAS Institute. 1988. SAS/STAT User’s Guide. 1988 Edition. SAS Inst. Inc., Cary, NC. Siegel, P. B. 1999. Utilizing resources for broilers. World Poult. 15:23–24. Shlosberg, A., M. Bellaiche, V. Hanji, A. Nyska, A. Lublin, M. Shemesh, L. Shore, S. Perk, and E. Berman. 1996. The effect of acetylsalicylic acid and cold stress on the susceptibility of broiler to the ascites syndrome. Avian Pathol. 25:581–590. Shlosberg, A., I. Zadikov, U. Bendheim, V. Handji, and E. Berman. 1992. The effects of poor ventilation, low temperatures, type of feed and sex on bird on the development of ascites in broilers. Physiopathological factors. Avian Pathol. 21:369–382. Vanhooser, S. L., A. Becker, and R. G. Teeter. 1995. Bronchodilator, oxygen level, and temperature effects on ascites incidence in broiler chickens. Poult. Sci. 74:1586–1590. Verstegen, M. W. A., A. M. Henken, W. Van der Hel, and M. T. Frankenhuis. 1989. Pages 65–69 in Proc. 7th Int. Conf. Prod. Dis. Farm Anim., Cornell Univ., Ithaca, NY. Wideman, R. F. 2000. Cardio-pulmonary hemodynamics and ascites in broiler chickens. Avian Poult. Biol. Rev. 11:21–43. Wideman, R. F., Jr., and W. G. Bottje. 1993. Current understanding of the ascites syndrome and future research directions. Pages 1–20 in Nutr. Tech. Symp. Proc. Novus Int. Inc., St. Louis, MO. Wideman, R. F., and G. F. Erf. 2002. Intravenous micro-particle injection and pulmonary hypertension in broiler chickens: Cardio-pulmonary hemodynamic responses. Poult. Sci. 81:877–886. Wideman, R. F., Jr., and H. French. 2000. Ascites resistance of progeny from broiler breeders selected for two generations using chronic unilateral pulmonary artery occlusion. Poult. Sci. 79:396–401. Wideman, R. F., Jr., and Y. K. Kirby. 1995. A pulmonary artery clamp model for inducing pulmonary hypertension syndrome (ascites) in broilers. Poult. Sci. 74:805–812. Wideman, R. F., Jr., Y. K. Kirby, R. L. Owen, and H. French. 1997. Chronic unilateral occlusion of an extrapulmonary primary bronchus induces pulmonary hypertension syndrome (ascites) in male and female broilers. Poult. Sci. 76:400–404. Witzel, D. A., W. E. Huff, L. F. Kubena, R. B. Harvey, and M. H. Elissalde. 1990. Ascites in growing broilers: A research model. Poult. Sci. 69:741–745.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Anthony, N. B., J. M. Balog, J. D. Hughes Jr., L. Stamps, M. A. Cooper, B. D. Kidd, X. Liu, G. R. Huff, W. E. Huff, and N. C. Rath. 2001. Genetic selection of broiler lines that differ in their ascites susceptibility 1. Selection under hypobaric conditions. Pages 327–328 in Proc. 13th Eur. Symp. Poult. Nutr., Blankenberge, Belgium. Balog, J. M. 2003. Ascites syndrome (pulmonary hypertension syndrome) in broiler chickens: Are we seeing the light at the end of the tunnel? Avian Poult. Biol. Rev. 14:99–126. Balog, J. M., N. B. Anthony, M. A. Cooper, B. D. Kidd, G. R. Huff, W. E. Huff, and N. C. Rath. 2000. Ascites syndrome and related pathologies in feed restricted broilers raised in a hypobaric chamber. Poult. Sci. 79:318–323. Beker, A., S. L. Vanhooser, J. H. Swartzlander, and R. G. Teeter. 2003. Graded atmospheric oxygen level effects on performance and ascites incidence in broilers. Poult. Sci. 82:1550– 1553. Boldman, K. L., A. Kriese, L. D. Van Vleck, and S. D. Kachman. 1993. A manual for the use of MTDFREML—A set of programs to obtain estimates of variances and covariances. ARS, USDA, Washington, DC. Buys, N., C. W. Scheele, C. Kwakernaak, and E. Decuypere. 1999. Performance and physiological variables in broiler chicken lines differing in susceptibility to the ascites syndrome: 2. Effect of ambient temperature on partial efficiencies of protein and fat retention and plasma hormone concentrations. Br. Poult. Sci. 40:140–144. Christensen, V. L., D. O. Noble, and K. E. Nestor. 2000. Influence of selection for increased body weight, egg production, and shank width on the length of the incubation period of turkeys. Poult. Sci. 79:613–618. De Greef, K. H., L. L. G. Janss, A. L. J. Vereijken, R. Pit, and C. L. M. Gerritsen. 2001. Disease-induced variability of genetic correlations: Ascites in broilers as a case study. J. Anim. Sci. 79:1723–1733. Dunnington, E. A., and P. B. Siegel. 1998. Resource allocations: Growth and immune responses. Pages 95–98 in Proc. 10th Eur. Poult. Conf. Falconer, D. S., and T. F. C. Mackay. 1996. Introduction to Quantitative Genetics. Pearson Educ. Ltd., Essex, UK. Julian, R. J. 1993. Ascites in poultry. Avian Pathol. 22:419–454. Julian, R. J. 1998. Rapid growth problems: Ascites and skeletal deformities in broilers. Poult. Sci. 77:1773–1780. Julian, R. J. 2000. Physiological, management and environmental triggers of the ascites syndrome: A review. Avian Pathol. 29:519–527. Julian, R. J., G. W. Friars, H. French, and M. Quinton. 1986. The relationship of right ventricular hypertrophy, right ventricular failure, and ascites to weight gain in broiler and roaster chickens. Avian Dis. 31:130–135. Julian, R. J., I. McMillan, and M. Quinton. 1989. The effect of cold and dietary energy on right ventricular hypertrophy, right ventricular failure and ascites in meat-type chickens. Avian Pathol. 18:675–684. Julian, R. J., and B. Wilson. 1992. Pen oxygen concentration and pulmonary hypertension induced right ventricular failure and ascites in meat type chickens at low altitude. Avian Dis. 36:733–735. Lister, S. 1997. Broiler ascites: A veterinary viewpoint. World’s Poult. Sci. 53:65–67.
2529
Key words: chicken, ascites, divergent selection, hypobaric hypoxia, heritability 2007 Poultry Science 86:2517–2529 doi:10.3382/ps.2007-00134
INTRODUCTION Intense selection for growth, feed conversion, and white meat yield in broilers has led to significant physiological changes. One of the changes of particular importance is the increased incidence of metabolic disorders, including ascites. Ascites is an economically important disease faced by the industry over the past 2 decades. It is a 2-fold problem, influencing both the live production and processing sectors of the industry. Typically, ascites mortality can range from 0 to 30% in broiler flocks. It has been estimated that 8% of 361,000,000 broiler deaths each year can be attributed to ascites. In 2002, it was estimated that
©2007 Poultry Science Association Inc. Received March 29, 2007. Accepted July 12, 2007. 1 Current address: Nicholas Turkey Breeding Farms, 31186 Midland Trail East, Lewisburg, WV 24901. 2 Current address: Cobb-Vantress, Inc., PO Box 1030, Siloam Springs, AR 72761. 3 Corresponding author: [email protected]
these on-farm late grow-out mortalities cost more than $100 million per year in the United States. In addition, 0.05% of all processing plant condemnations can be attributed to ascites (A. Rossi, Cobb-Vantress Inc., Siloam Springs, AR, personal communication). Ascites is a metabolic disease characterized by an enlarged flaccid heart, variable liver changes, and accumulation of fluid in the abdominal cavity (Riddell, 1991). Ascites is a result of the inability of the broiler to provide its tissues with an adequate supply of oxygen (Lister, 1997). The ascites syndrome cannot be classified as a contagious or infectious disease, but rather as a progressive disease starting with pulmonary hypertension and progressing to right side congestive heart failure, with the end result of congestive heart failure (Lister, 1997; Mitchell, 1997). Several management techniques have been developed and implemented to help curb the incidence of ascites in broiler flocks. Unfortunately, these techniques are designed to slow early bird growth, thus not allowing the bird to achieve its full genetic potential. Feed restriction is one the techniques most commonly used to curb the
2517
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
ABSTRACT Chicken lines that were either resistant or susceptible to ascites syndrome were developed by using a hypobaric chamber to induce the disease. Birds were reared in a hypobaric chamber that simulated high altitude by operating under a partial vacuum, which thereby lowered the partial pressure of oxygen. Ascites mortality data from birds reared under hypobaric chamber conditions were used to select siblings to be used for breeding. The response to selection for the susceptible (SUS) and resistant (RES) lines of chickens was very rapid from the base population, which exhibited an incidence of ascites of 75.3%. Extremes in the incidence of ascites were observed in generation 8, with line SUS exhibited an average incidence of ascites of 95.1%, and in generation 9, with line RES exhibited an average incidence of ascites of 7.1%. The incidence of ascites in the relaxed line remained relatively stable and currently has a general incidence of ascites of 60%. The heritability estimates ± SE for ascites were estimated to be 0.30 ± 0.05 and 0.55 ± 0.05 for lines SUS and RES, respectively. Changes in the incidence of
2518
PAVLIDIS ET AL.
ascites and lived to sexual maturity were chosen to be the first generation in an ascites-resistant line. Successive generations were reared under cold stress conditions to screen for ascites susceptibility. After 2 generations, progeny from this RES line had significantly lower ascites mortality compared with the base population. The males exhibited an ascites mortality of 6.4% and the females 0% compared with the base populations, which exhibited an ascites mortality of 43.6 and 12.3% respectively. This unidirectional selection study shows that ascites resistance can be successfully selected against. However, in the study by Wideman and French (2000), valuable information such as heritabilities and genetic correlations were not estimated and correlated responses were not measured. Divergent selection for ascites has been practiced in the authors’ laboratory through the use of a hypobaric chamber, which simulates high-altitude conditions, to repeatably induce ascites, thereby allowing exploration of the genetic relationship and correlated responses associated with selection for ascites. The purpose of the current research was to summarize the results of 10 generations of selection for ascites.
MATERIALS AND METHODS Hypobaric Chamber The hypobaric and local altitude chambers were each 2.4 × 3.7 × 2.4 m, and each was capable of housing 240 broilers to 6 wk of age. The chambers were equipped with identical custom stainless steel batteries fitted with trough feeders and nipple waterers. The chambers were matched in terms of ventilation rate, temperature, and size. The environmental variables monitored included altitude, ventilation, temperature, and humidity. Ventilation was set to maintain a constant airflow at the rate of 17 m3/min. Birds were warm room brooded, with temperatures decreased weekly. The hypobaric chamber was set to simulate 2,900 m (9,500 ft) above sea level. A control environment was maintained in a local altitude chamber at 390 m (1,200 ft) above sea level. All birds placed at simulated high altitude remained at high altitude for the duration of the experiments. Daily management tasks and weighings were conducted in the respective rearing environment.
Initial Line Formation Figure 1 illustrates the development of the ascites-susceptible (SUS) and ascites-resistant (RES) lines from a commercial pedigree elite line (REL). In 1995, 3 separate hatches of pedigree male line chicks were obtained from a primary poultry breeding company. The chicks obtained in these hatches represented offspring from 16 sire families in the pedigree male line. This pedigree male line had undergone one generation of relaxed selection prior to becoming the base population for development of the ascites lines. The chicks obtained in the first and third hatches were reared under typical broiler management conditions in the
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
incidence of ascites (Julian, 1998; Balog et al., 2000; Roush and Wideman, 2000; Wideman, 2000; Balog, 2003). Another technique used is intermittent lighting (Julian, 2000). Ascites had been a difficult syndrome to study because of the problems associated with obtaining a uniform characteristic flock incidence. Thus, to better understand both the physiological and genetic components, a repeatable model needed to be developed. Both surgical and nonsurgical methods have been developed. Surgical methods include clamping of the left pulmonary artery (Wideman and Kirby, 1995) and unilateral occlusion of the primary bronchus (Wideman et al., 1997). Wideman and Erf (2002) reported that the intravenous injection of microparticles can also be an effective method of inducing ascites. Nonsurgical methods of inducing ascites include various cold stress models, which involve exposing the bird to constant cool temperatures (Julian et al., 1989; Vanhooser et al., 1995), to a gradual decrease in temperature (Verstegen et al., 1989; Buys et al., 1999), or to an episodic cold stress (Shlosberg et al., 1996). Low-ventilation models, which simulate a winterized broiler house, have also been implemented, but this method does not induce ascites to the same magnitude as cold stress (Julian and Wilson, 1992; Shlosberg et al., 1992). High-altitude simulation through the use of a hypobaric chamber is another nonsurgical model producing a consistent incidence of ascites in broilers (Owen et al., 1990; Witzel et al., 1990; Balog et al., 2000; Anthony et al., 2001). By operating under a partial vacuum, a reduction of the partial pressure of oxygen is created, thus simulating conditions observed in naturally high altitudes. Foundation experiments involving the hypobaric chamber allowed only a small number of individuals to be tested (Owen et al., 1990). The use of larger hypobaric chambers with tight environmental conditions has increased this model’s usefulness (Balog et al., 2000; Anthony et al., 2001). In addition, constant monitoring of environmental variables such as altitude, ventilation, temperature, and humidity provides the researcher more environmental control. This allows the researcher the ability to repeatedly “dial in” the same environment every time the chamber is used, thus making it one of the most ideal noninvasive models for inducing ascites in broilers. Ascites is known to be influenced by both environmental (Julian, 2000) and genetic factors (Lubritz et al., 1995; Wideman and French, 2000; Anthony et al., 2001) and is an artifact of intense genetic selection for rapid growth rate, feed conversion, heavy BW, and white meat yield in broiler populations. Lubritz et al. (1995) used cold stress as the model to induce ascites in 3 commercial male lines and reported moderate to high heritabilities (0.11 to 0.44) for ascites score, and moderate heritabilities (0.21 to 0.27) for the ratio of the right ventricle (RV) to total heart weight. Therefore, it seemed likely that a selection program could be developed and be successfully implemented to reduce the incidence of ascites in broiler populations. Wideman and French (2000) used the pulmonary artery clamp to screen a small population of broilers for ascites susceptibility. The individuals that did not succumb to
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
2519
hypobaric and local altitude chambers. Birds reared in the hypobaric chamber were used to generate ascites mortality data, which was used to make sib selection decisions. Birds reared in the local altitude chamber served as a control for the altitude challenge. All birds were provided ad libitum consumption of water and a corn and soybean-based broiler starter and grower ration that was formulated to meet or exceed NRC (1994) requirements. The photoperiod consisted of 23 h of light and 1 h of dark. All the hypobaric chamber daily management tasks and weighings were conducted under partial vacuum through the use of an air lock, which allowed for pressure equilibrium. Growth was evaluated by measuring BW at d 21 and 42 (BW21 and BW42, respectively) on all survivors. Birds were checked daily and all mortalities were necropsied to determine the cause of death. At necropsy, mortalities were classified as ascitic or nonascitic. Mortalities were evaluated for symptoms of ascites, such an enlarged and flaccid heart, liver lesions, presence of fluid in the pericardium, and presence of fluid in the abdominal cavity. Mortalities were classified as ascitic if fluid was present in the abdominal cavity. The BW at death, sex, and day of death in the study were also recorded at the time of necropsy. At 6 wk, all remaining survivors in both the local and hypobaric chambers were euthanized by cervical dislocation and necropsied to check for the presence of ascites. Individual hearts, livers, and spleens were removed and individual organ weights recorded. The hearts were dissected into the RV and the remaining heart, and RV were
individually weighed to calculate the RV to total ventricle (TV) ratio (RV:TV ratio), which has been shown to be an indicator of ascites (Julian, 1993; Wideman and Bottje, 1993; Lubritz et al., 1995). Birds were considered ascitic if fluid was present in the abdominal cavity. Sire family selection was applied to develop the SUS and RES lines. Average ascites mortality was calculated for each of the original 16 sire families. The 8 sire families with the lowest average ascites mortality percentages were selected to become the RES line, and the 8 sire families with the highest average ascites mortality percentages became the SUS line. The REL line was maintained as a control and is currently maintained as a randomly mated population with the stipulation that full- and half-sib matings are not permitted. At the same time that chicks from the first and third hatches were being challenged in the hypobaric chamber, chicks obtained from the second hatch were placed in floor pens and grown as pedigreed broilers. These chicks were grown under typical broiler breeder management conditions, as suggested by the primary breeder from which they originated. Up to 5 wk of age, birds were allowed ad libitum access to a corn and soybean-based broiler ration that was formulated to meet or exceed NRC (1994) requirements. At 5 wk, individual BW were obtained; birds were sex separated and placed on a feed and photoperiod restriction program to delay the onset of lay. Water was available for ad libitum consumption throughout the entire rearing period.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 1. Development of the ascites-susceptible and ascites-resistant lines for a commercial pedigree elite line based on the performance of progeny when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level.
2520
PAVLIDIS ET AL.
At 20 wk of age, average ascites mortality data obtained in the altitude challenges were used to randomly select 3 males from each of the 8 sire families from the SUS and RES lines. This resulted in the creation of 24 sire families within each of the respective lines. Each selected male was housed in individual broiler breeder male cages and randomly assigned to 3 randomly chosen females from the same line, with the stipulation that full-sib and halfsib matings would not be permitted. Matings were performed by artificial insemination, and complete pedigree information was collected. The respective lines were closed after their initial formation.
Current Selection Methods The method of selection and propagation of the ascites lines is illustrated in Figure 2. All matings were performed by artificial insemination, and complete pedigree information was collected. Each generation, 2 hatches of offspring were challenged at high altitude in the hypobaric chamber. Hypobaric mortality information was used to select sibs grown under normal industry conditions. These pedigree replacements were derived from 2 to 4 pedigree hatches from the same sire families that provided the hypobarically challenged chicks. The floor-reared pedigree replacements were maintained under typical broiler breeder management condi-
tions as described during initial line formation. Briefly, birds were allowed ad libitum access to an NRC (1994)formulated broiler ration up to 5 wk of age. At 5 wk, individual BW were obtained, and birds were separated by sex and placed on a feed and photoperiod restriction program to control growth and promote optimal reproductive performance later in life. Water was available for ad libitum consumption throughout the entire rearing process. For the altitude challenge, birds were reared under typical broiler management conditions and allowed ad libitum access to water and an NRC (1994)-formulated corn and soybean-based starter and grower ration. The photoperiod consisted of 23 h of light and 1 h of dark. All the hypobaric chamber daily management tasks and weighings were conducted under partial vacuum through the use of an air lock, which allowed for pressure equilibrium. Growth was evaluated by measuring the BW of all survivors at 3 and 6 wk. At 6 wk, all remaining survivors were euthanized by cervical dislocation and necropsied to check for the presence of ascites as described previously. From the altitude challenge, average ascites mortality for each sire family was calculated and used to select breeders from the floor-reared sibs. For each line, breeder males were randomly selected from no more than the top 6 sire families, and females from no more than the top 10 sire families. The selected breeders were placed in individual
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 2. Selection schematic for the ascites-susceptible and ascites-resistant lines by using hypobaric exposure in a hypobaric chamber set at 2,900 m above sea level as the method of inducing ascites.
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
2521
cages between 18 and 20 wk of age. A sire family for each line was constructed by randomly assigning 3 females to each of 24 males with the stipulation that full-sib and halfsib matings were not permitted.
Data Analysis The change in traits over generations for lines SUS, RES, and REL were estimated by linear regression of line means on generations by using the procedure regression feature in SAS software (SAS Institute, 1988). Data were analyzed according to the multitrait derivative free restricted maximum likelihood procedures outlined by Boldman et al. (1993). An animal model was used in both lines SUS and RES to estimate heritabilities and genetic correlations. For ascites, RV:TV ratio, and livability, no fixed effects or covariates were used. For BW21 and BW42, ascites was used as a covariate and sex was used as a fixed effect. Generation was not used in any model to estimate heritabilities or genetic correlations in either line (L. D. Van Vleck, Animal Science Department, University of Nebraska, Lincoln, personal communication).
RESULTS AND DISCUSSION Ascites Mortality The changes per generation for the incidence of ascites in lines SUS, RES, and REL are presented in Figure 3. Significant linear regression coefficients were observed in all 3 lines for the percentage of ascites mortality. For line SUS, there was a significant average increase in ascites mortality of 0.95% per generation, whereas in line RES
there was a significant decrease in ascites mortality of 6.75% per generation. Although no selection pressure was applied to line REL for any traits, a significant reduction in ascites mortality of 3.14% per generation was observed. The response to selection for the ascites SUS and RES lines of chickens was very rapid, with divergence for ascites mortality beginning to occur in generation 3. To date, extremes in the incidence of ascites have been observed in generation 8 for line SUS (95.14%) and in generation 9 for line RES (7.14%). The incidence of ascites in line REL has decreased slightly; however, it has had an overall average incidence of approximately 60%. The response to selection for incidence of ascites in lines SUS and RES was extremely rapid, suggesting that a large genetic component may be associated with ascites mortality. Therefore, the heritability for ascites was expected to be high. Heritability estimates for ascites mortality in lines SUS and RES are presented in Tables 1 and 2, respectively, and as expected, were found to be high. Lubritz et al. (1995) were the first to report heritabilities for ascites in birds reared under cold stress conditions, and these heritabilities ranged from 0.11 to 0.44, depending on the genetic line tested. Other studies (De Greef et al., 2001; Pakdel et al., 2002) in which birds were reared under cold stress have reported heritability estimates consistent with those reported by Lubritz et al. (1995). The heritability estimates reported in the current study are consistent with those reported for birds reared under cold stress conditions (Lubritz et al., 1995; De Greef et al., 2001; Pakdel et al., 2002). The heritability for ascites mortality was higher in line RES than in line SUS (Tables 1 and 2). These differences in heritabilities can be attributed to the differences in response to selection for ascites mortality. The response to
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 3. Incidence of ascites mortality per generation for lines susceptible (SUS) and resistant (RES) to ascites and for a nonselected control line (REL) when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level.
2522
PAVLIDIS ET AL. Table 1. Estimates of heritabilities, genotypic correlations, and phenotypic correlations of traits measured under hypobaric conditions in the ascites-susceptible line of chickens1 Trait Ascites RV:TV Livability BW21 BW42
Ascites
RV:TV
Livability
BW21
BW42
0.30 (0.05) NC2 0.73 0.28 0.18
−0.27 0.35 (0.10) 0.00 NC −0.23
0.87 −0.03 0.47 (0.05) 0.64 −0.06
0.55 NC 0.58 0.66 (0.07) 0.71
0.71 0.19 0.43 0.71 0.55 (0.12)
1 Genetic correlations are above the diagonal, phenotypic correlations are below the diagonal, and heritabilities (SE) are on the diagonal. RV:TV = right ventricle:total ventricle weight; BW21 = BW at d 21; BW42 = BW at d 42. 2 NC = no convergence obtained.
Livability The change per generation for livability in lines SUS, RES, and REL are presented in Figure 4. Livability is defined as how long a bird is able to live in the hypobaric chamber during a 42-d period without succumbing to ascites. Significant linear regression coefficients were observed for all lines for livability. Over 10 generations, selection in line RES has extended livability by 11.5 d, whereas in line SUS, livability has been decreased by 8 d. These changes in livability in lines RES and SUS appear to be associated with the timing of the onset of ascites for the respective lines. Figure 5 clearly shows that selection for ascites has shifted the ascites mortality curves, resulting in line SUS succumbing to ascites earlier than line RES. Line SUS typically has ascites mortalities beginning at approximately d 5, whereas line RES does not have ascites mortalities until approximately d 11. On average, birds in
line REL had an average day of death from ascites at 27 d, whereas lines SUS and RES were at 19 and 22 d, respectively. Therefore, selection in both lines has resulted in birds succumbing to ascites earlier than those in line REL. This is especially beneficial in line RES, because ascites mortalities that occur are relatively early in the growout period, when resources such as feed and economic loss are minimal. However in line REL, the average livability was 29 d, whereas livability in line SUS was decreased to 22 d and was increased in line RES to 39 d. These changes in livability are directly related to the respective ascites mortalities observed for lines SUS and RES (Figure 3). The response in livability was extremely rapid, and livability appears to be correlated with ascites mortality in both lines SUS and RES. Therefore, the heritability for livability was expected to be high, and positive genetic and phenotypic correlations were expected between ascites mortality and livability. The heritability estimates for livability as well as the genetic and phenotypic correlations between ascites mortality and livability are presented in Tables 1 and 2. Heritabilities were found to be high, and there were strong, positive genetic and phenotypic correlations between ascites incidence and livability in lines SUS and RES. Therefore, selection for ascites susceptibility will result in a reduction in livability, whereas selection for ascites resistance will increase livability.
BW at Hatch, d 21, and d 42 The relationship between ascites and BW has been well documented (Julian et al., 1986; Julian, 2000; Balog, 2003). Birds that have heavier BW and exhibit faster growth rates
Table 2. Estimates of heritabilities, genotypic correlations, and phenotypic correlations of traits measured under hypobaric conditions in the ascites-resistant line of chickens1 Trait Ascites RV:TV Livability BW21 BW42
Ascites
RV:TV
Livability
BW 21
BW 42
0.55 (0.05) 0.25 0.81 0.24 −0.05
−0.047 0.59 (0.06) 0.52 NC −0.08
0.92 0.19 0.70 (0.5) 0.46 −0.07
0.20 NC2 0.37 0.89 (0.05) 0.62
−0.23 0.13 −0.25 0.72 0.56 (0.07)
1 Genetic correlations are above the diagonal, phenotypic correlations are below the diagonal, and heritabilities (SE) are on the diagonal. RV:TV = right ventricle:total ventricle weight; BW21 = BW at d21; BW42 = BW at d 42. 2 NC = no convergence obtained.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
selection for ascites mortality was greater in line RES than in line SUS. This difference in the response for ascites mortality may be because line RES has a greater range to reduce ascites mortality than the range line SUS has to increase ascites mortality, because ascites mortality was 75.3% in generation 0. Obviously, ascites mortality cannot exceed greater than 100% for line SUS or less than 0% for line RES. With an average incidence of ascites of 95.1%, line SUS began to reach a physiological ceiling by generation 8 (Figure 3). Approaching this physiological ceiling limits the amount of progress that can be made in this line; thus, heritability for the trait would begin to decrease, consistent with what was seen with the heritability estimates for line SUS as compared with line RES (Tables 1 and 2).
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
2523
are more likely to develop ascites than those birds with lighter BW and slower growth rates (Julian et al., 1986; Julian, 2000; Balog, 2003). Because there is a defined negative relationship between ascites and BW, it is important to understand how selection for ascites susceptibility and resistance influences BW, a trait of high economic importance to the industry. The effects of selection on hatch weight, BW21, and BW42 in both the hypobaric chamber and at sea level are presented in the various panels of Figure 6. For hatch weight (Figure 6A), significant negative linear regression coefficients were observed in lines SUS and RES; however,
no significant change was observed for line REL. This resulted in line RES producing chicks that were slightly heavier than those in line SUS at hatch. However, when the initial 16 sire families were divided into lines SUS and RES in generation 0, it was unclear whether selection for susceptibility to ascites had resulted directly in a reduction in hatch weight in line SUS or whether these differences in hatch weight could be attributed to a potential founder effect (Falconer and Mackay, 1996). Another possible explanation for the differences seen in hatch weight could be the impact of selection on incubation time. Christensen et al. (2000) reported the effects of selection in various lines
Figure 5. Typical ascites mortality curve for lines susceptible (SUS) and resistant (RES) to ascites and for a nonselected control line (REL) when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 4. Livability per generation for lines susceptible (SUS) and resistant (RES) to ascites and for a nonselected control line (REL) when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level.
2524
PAVLIDIS ET AL.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 6. Change per generation for hatch weight, BW at 21 d (BW21), and BW at 42 d (BW42) for lines susceptible (SUS) and resistant (RES) to ascites and for a nonselected control line (REL) reared under normal environmental conditions in the sea level chamber and when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level. A) Hatch weight per generation for ascites lines SUS and RES and for the nonselected control line REL under normal environmental conditions. B) BW21 per generation for ascites lines SUS and RES and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. C) BW42 per generation for ascites lines SUS and RES and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. D) BW21 per generation for ascites lines SUS and RES and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber. E) BW42 per generation for ascites lines SUS and RES and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber.
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
died from ascites, whereas a smaller number had died from ascites in line RES. Thus, those birds in line SUS that had survived the 42-d challenge in the hypobaric chamber could be considered the least susceptible birds in line SUS. For those birds in line SUS to survive the hypobaric challenge, they would have needed growth rates and final BW that were similar to those of line RES, although after 10 generations, the survivors in line SUS had BW that were approximately 32 g heavier than the survivors of line RES. The impact on BW42 in line REL is approximately half that seen in lines SUS and RES. Because the incidence of ascites seen in line REL was intermediate to those seen in lines SUS and RES, the impact on BW42 for line REL in the hypobaric chamber would be expected to be intermediate to the effects seen for lines SUS and RES. For BW42 at sea level (Figure 6C), significant negative linear regression coefficients were observed across all lines. Line SUS did maintain a superior BW across both lines RES and REL. This superiority in BW for line SUS was supported by both the high, positive genetic and phenotypic correlations between ascites and BW42 (Table 1). By generation 10, line SUS had lost approximately 214 g in BW42, whereas line RES had lost 377 g. This resulted in line SUS having BW that were approximately 163 g heavier than those of line RES after 10 generations of selection. This reduction in BW in line RES was consistent with the negative genetic and phenotypic correlations reported between ascites and BW42 for line RES (Table 2). The superiority in BW for line SUS combined with the positive genetic and phenotypic correlations were consistent with the relationships that have been reported between ascites susceptibility and BW (Julian et al., 1986; Lubritz et al., 1995; Julian, 2000; Pakdel et al., 2002; Balog, 2003). Birds with heavier BW at d 21 typically had higher RV:TV ratios and were more likely to develop ascites than those birds that had lighter BW at the same age (Julian et al., 1986). Under normal environmental conditions, the increase in BW21 observed for line SUS and the decrease seen in line RES were consistent with the findings of Julian et al. (1986). Although significant negative linear regression coefficients were observed for all lines for BW42 under normal conditions, line SUS did maintain superiority for BW42 over both lines REL and RES (Figure 6C). When BW42 in line SUS was expressed as a deviation from the base population line REL, positive BW gains were detected (data not shown). Beker et al. (2003) reported unpublished data from Teeter and Wiernusz that stated the oxygen requirements for 1 g of lean and of fat accretion were 3.9 and 1.2 L, respectively, to 42 d of age in commercial broilers from 1994. We hypothesized that, in our laboratory, the differences in BW21 and BW42 seen in the current study between lines SUS and RES under normal conditions were due mainly to differences in lean tissue mass as opposed to large differences in fat mass. Under this hypothesis, the demand for oxygen would be increased in line SUS and decreased in line RES. Therefore, when reared under normal environmental conditions, when oxygen was not a limiting factor, growth was optimized, resulting in line
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
of turkeys and their respective randombred control lines, and found that selection had modified incubation time. From those data, one could hypothesize that line RES had delayed hatching, thus resulting in wetter chicks that would be heavier than earlier hatching chicks from line SUS. Although this may be a potential explanation for the differences in hatch weight between lines SUS and RES, further research is needed to determine why this difference is present. In the hypobaric chamber for BW21 (Figure 6D), significant negative linear regression coefficients were observed for lines SUS and RES, with line SUS having a larger reduction in BW per generation than line RES. No significant change was observed in BW21 for line REL. By generation 10 in line RES, there was a reduction in BW of approximately 47 g; this reduction in BW mimics current management strategies designed to minimize the incidence of ascites. These management strategies are designed to slow early growth and allow for the development of a cardiopulmonary system that is capable of supporting rapid growth while minimizing the impact on final BW (Julian, 1998, 2000; Balog et al., 2000; Roush and Wideman, 2000; Balog, 2003). By generation 10, line SUS was approximately 29 g lighter than line RES. One must remember that by d 21, more than 50% of the birds in line SUS had died from ascites, and a larger portion of the remaining birds were in various stages of progression of ascites than in line RES. Because a larger portion of birds in line SUS were having to deal with ascites compared with those in line RES, more resources in line SUS were being directed toward maintenance to try to cope with ascites. This reallocation of resources toward maintenance resulted in a reduction in the amount of resources being devoted to growth (Dunnington and Siegel, 1998; Siegel, 1999) and could be a potential explanation for the higher reduction in BW seen in line SUS compared with line RES. For BW21 at sea level (Figure 6B), significant linear regression coefficients were observed for all lines. Gains in BW per generation were observed for lines SUS and REL, whereas there was a reduction in BW for line RES. By generation 10, line SUS was approximately 112 g heavier than line RES. Julian et al. (1986) reported a positive relationship between ascites and BW at d 21 and 28, suggesting that selection for increased susceptibility to ascites would result in an increase in BW21. Again, as seen with line RES in the hypobaric chamber (Figure 6D), under normal conditions in which oxygen is not a limiting factor, there was a significant reduction in BW. This reduction in early BW under normal conditions further supports the fact that selection for ascites resistance has resulted in a pattern of growth that mimics current industry practices for management of ascites. For BW42 (Figure 6E) in the hypobaric chamber, significant negative linear regression coefficients were observed across all lines. Selection had the greatest impact on BW in line RES, although line SUS had final BW42 that were similar to those of line RES. This similarity in final BW between lines RES and SUS was not unexpected, because by d 42 the majority of the birds placed in line SUS had
2525
2526
PAVLIDIS ET AL.
Heart Variables The impact of selection for resistance and susceptibility to ascites on the right and left ventricles relative to BW, as well as its impact on the RV:TV ratio in both sea level and hypobaric conditions, are presented in the various panels of Figure 7. Across all lines for birds reared at sea level, no significant changes were observed for relative RV (Figure 7B). For relative left ventricle, no significant changes were observed for lines SUS and RES; however, there was a slight reduction in line REL (Figure 7C). For those birds reared in the hypobaric chamber, significant linear regression coefficients were observed for relative RV for lines SUS and RES, whereas no significant changes were observed in line REL (Figure 7E). The reduction per generation for relative RV in line RES was 3 times greater than that observed in line SUS (Figure 7E). For relative left ventricle, significant negative linear regression coefficients were observed across all lines, and the change in relative left ventricle in line SUS was 3 times greater than that of line RES (Figure 7F).
Because the available oxygen is significantly reduced in hypobaric conditions, the natural response is to increase red blood cell production to increase the oxygen-carrying capacity (Balog et al., 2000). This increase in red blood cell production results in an increase in blood hematocrit and, in turn, causes an increase in blood viscosity (Balog et al., 2000). This increase in blood viscosity results in the heart having to work harder to pump this viscous blood to the tissues, resulting in ventricular hypertrophy, more specifically, RV hypertrophy (Julian, 2000; Wideman 2000; Balog, 2003). As the demand for oxygen increases, as seen in rapidly growing broilers, ventricular hypertrophy is greater than that of slower growing birds. This makes the faster growing birds more likely to succumb to heart failure and eventually death (Julian, 2000; Wideman, 2000; Balog, 2003). From the data in the present selection experiment, selection for resistance and susceptibility to ascites under normal conditions did not result in a significant impact on the relative right and left ventricles; however, it did result in a disproportionate change in the right and left ventricles in lines SUS and RES under hypobaric conditions. These disproportionate changes in the right and left ventricles under hypobaric conditions in lines SUS and RES are important in understanding the physiological changes that occur in the heart and how it is related to the bird’s ability to survive an ascites challenge, and thus require further investigation. At sea level, no significant changes were observed for the RV:TV ratio for lines SUS and RES (Figure 7A). Because no significant changes were observed for the relative right and left ventricles in lines SUS and RES (Figures 7B and 7C), no changes in the RV:TV ratio were expected (Figure 7A) for these lines. However, there was a slight reduction in the RV:TV ratio for line REL (Figure 7A). In the hypobaric chamber, significant negative linear regression coefficients were observed for all lines, with line RES having the greatest reduction in RV:TV ratio over 10 generations of selection (Figure 7D). The RV:TV ratio has been widely used as an indicator trait for ascites (Wideman and Bottje, 1993; Julian, 2000; Wideman, 2000). An RV:TV ratio of greater than 0.27 has been used as a cutoff point for determining the ascites status of a bird, with birds with RV:TV ratios greater than 0.27 being considered ascitic (Wideman and Bottje, 1993; Julian, 2000; Wideman, 2000). Positive genetic and phenotypic correlations between ascites and RV:TV ratio for birds reared under cold stress has been reported, and these correlations suggest that selection for decreased ascites mortality will result in a reduction in the RV:TV ratio (Lubritz et al., 1995; Pakdel et al., 2002). Although a low but positive phenotypic correlation for ascites and RV:TV ratio was observed in line RES (Table 2), under normal conditions no significant changes for the RV:TV ratio were observed in the selected lines (Figure 7A). However, line SUS had mean RV:TV ratio values that were greater than those of line RES (Figure 7A). This lack of change in RV:TV ratio for the lines was not consistent with what has been hypothesized regarding the impact of selection for ascites on the RV:TV ratio (Lubritz et al., 1995, Pakdel et al., 2002).
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
SUS having BW21 and BW42 that were greater than that of line RES. Under this hypothesis, the potential differences between lines SUS and RES in oxygen demand for growth were further exacerbated when these lines were reared in the hypobaric chamber, where oxygen was the limiting factor. However, when these lines were placed in the hypobaric chamber, where oxygen was a limiting factor, line RES, through its reduction in BW, specifically early BW, had in turn a reduced demand for oxygen and was better able to handle the challenge of the oxygen-limiting hypobaric chamber. Line SUS, which had increased BW, and in turn, increased oxygen demand to support the increase in BW observed at d 21 and 42, was unable to meet the demand because of limited oxygen availability, which led to a cascade of events, ultimately ending in death from ascites. This hypothesis regarding the impact of selection on BW and its impact on the demand for oxygen is consistent not only with the observed ascites mortality (Figure 3), but also with livability (Figure 4) and the timing of when birds succumbed to ascites in the hypobaric chamber (Figure 5). These reductions in BW21 and BW42 in line RES were consistent with current management practices, which are designed to slow early growth of the bird to allow the cardiopulmonary system to develop, often at the cost of the genetic potential of the birds. This reduction in BW for line RES has potential economic implications that could limit integrating selection for ascites resistance into a breeding program. Therefore, a geneticist must determine whether the benefit of having resistance to ascites outweighs the loss in economically important BW, and potentially other traits that may be related to ascites. For selection for ascites resistance to be economically beneficial, selection for BW must occur simultaneously. However, the addition of BW selection and any other production traits will reduce the progress that is made for resistance to ascites.
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
2527
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Figure 7. Change per generation for the right ventricle:total ventricle (RV:TV) ratio, right ventricle (RV) relative to BW at 42 d (BW42), and left ventricle relative to BW42 for lines susceptible (SUS) and resistant (RES) to ascites and a nonselected control line (REL) reared under normal environmental conditions in the sea level chamber and when reared under hypobaric conditions in the hypobaric chamber set at 2,900 m above sea level. A) RV:TV ratio per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. B) RV relative to BW42 per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. C) Left ventricle relative to BW42 per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under normal environmental conditions in the sea level chamber. D) RV:TV ratio per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber. E) RV relative to BW42 per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber. F) Left ventricle relative to BW42 per generation for lines SUS and RES to ascites and for the nonselected control line REL when reared under hypobaric conditions in the hypobaric chamber.
Although no significant differences were observed for the change in the RV:TV ratio in lines SUS or RES (Figure 7A), the differences in mean RV:TV values could be a result of differences between the initial 16 sire families for the
RV:TV ratio when they were split into lines SUS and RES. However, under normal conditions, none of these mean values exceeded the 0.27 threshold once divergence for ascites mortality had occurred (Figure 7A).
2528
PAVLIDIS ET AL.
The current data, combined with the negative genetic correlations and low positive phenotypic correlations reported in the current study between ascites and RV:TV ratio for lines SUS and RES (Tables 1 and 2), suggested that selection for resistance and susceptibility to ascites has not directly modified the RV:TV ratio. Although heritability of the RV:TV ratio was high, the phenotypic correlations, combined with the negative genetic correlations between the RV:TV ratio and ascites (Tables 1 and 2), suggested that sole selection for the RV:TV ratio may not be the best means for reducing the incidence of ascites in pedigree populations.
Inbreeding
General Synthesis The response to selection for the SUS and RES lines of chickens was very rapid from the base population, which exhibited an incidence of ascites of 75.30% (Figure 3). To date, extremes in the incidence of ascites were observed in generation 8 for line SUS (95.14%) and in generation 9 for line RES (7.14%; Figure 3). The heritabilities for ascites were estimated to be 0.30 ± 0.05 and 0.55 ± 0.05 for lines SUS and RES, respectively (Tables 1 and 2). These heritability estimates are consistent with what has previously been reported for ascites measured under cold stress conditions (Lubritz et al., 1995; De Greef et al., 2001; Pakdel et al., 2002). The rapid response seen for the incidence of ascites and the moderate to high heritabilities suggest that a few major genes may control ascites. Selection for resistance and susceptibility to ascites seems to be associated with livability. By generation 10, selection for ascites in line RES had increased livability by 11.5 d, whereas in line SUS, livability had been decreased by 8 d (Figure 4). Selection for ascites susceptibility also decreased the number of days it took for birds in line SUS to succumb from ascites, whereas in line RES it increased the number of days (Figure 5). Previously reported positive genetic correlations between ascites and the RV:TV ratio suggest that selection for a decreased RV:TV ratio will result in a reduction in the incidence of ascites (Lubritz et al., 1995). However, in the current study negative genetic correlations between ascites and the RV:TV ratio were observed in both lines SUS and RES. In addition, after 10 generations there has been no significant change in the RV:TV ratio in lines SUS and RES under normal conditions. The current data raise questions about the validity of using the RV:TV ratio as the sole indicator trait in a selection program designed to reduce the incidence of ascites. Under normal conditions, selection for ascites susceptibility and resistance has resulted in modifications of both
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Inbreeding was calculated by using pedigree information from the floor-reared sibs in an animal model (Boldman et al., 1993). The average changes in inbreeding per generation were 5.01, 5.12, and 2.14%, respectively, for lines SUS, RES, and REL.
BW21 and BW42 for lines SUS and RES (Figure 6). The modification of BW21 (reduced in line RES and increased in line SUS) was consistent with the relationship between BW21, RV:TV ratio, and ascites mortality reported by Julian et al. (1986). Although significant negative linear regression coefficients were observed for all lines under normal environmental conditions for BW42, line SUS did maintain superiority in BW42 over lines RES and REL. Under normal conditions, if BW42 in line SUS were to be expressed as a deviation from the base population line REL, then positive gains for BW42 would be observed. This increase in BW42 in line SUS under normal conditions was in agreement with the previously documented relationships between ascites mortality and BW (Julian et al., 1986; Julian, 2000; Balog, 2003; Pakdel et al., 2005). The reduction in both BW21 and BW42 seen in line RES when reared under normal conditions mimics current industry practices for minimizing the incidence of ascites. Current industry practices are designed to slow early growth to allow a sufficient cardiopulmonary system to develop, often resulting in the bird being unable to maximize its genetic potential (Julian, 1998; Balog et al., 2000; Roush and Wideman, 2000; Balog, 2003). The reduction in BW21 and BW42 in line RES was clearly related to both ascites mortality and livability (Figures 3 and 4), as well as the timing of when birds succumb to ascites in the limited oxygen environment of the hypobaric chamber (Figure 5). Overall, direct selection for resistance to ascites by using sire family performance appeared to be an effective means of reducing the incidence of ascites in broilers. However, selection for resistance to ascites did result in a reduction in both BW21 and BW42 when the birds were reared under normal environmental conditions. Therefore, a geneticist must determine whether the benefits of having resistance to ascites outweigh the loss seen in BW and the potential loss in other economically important traits. Pakdel et al. (2005) reported that selection that included ascites-related traits such as hematocrit and RV:TV ratio, when measured under both normal and cold stress conditions, would result in an increase in BW but in no genetic change for ascites mortality. This result suggested that current levels of ascites mortality in breeding populations could be maintained without loss of BW (Pakdel et al., 2005). From the data reported in the current selection program, selection to reduce ascites mortality clearly came at the cost of reducing BW. Therefore, to make genetic improvement in a breeding population for ascites mortality without a loss in BW, concomitant selection for BW and other production parameters would have to be performed. One will have to determine breeding goals while realizing that the amount of selection pressure applied to these traits will directly influence the amount of progress that is made for ascites resistance and BW. In the current study, only the generational response to selection for ascites susceptibility and resistance on various heart parameters and BW measurements is reported. Further research needs to be completed to gain a better understanding of how selection for susceptibility and resistance to ascites influences other economically important produc-
DIVERGENT SELECTION FOR ASCITES INCIDENCE IN CHICKENS
tion traits, such as growth, feed conversion, meat yield, and quality.
REFERENCES
Lubritz, D. L., J. L. Smith, and B. N. McPherson. 1995. Heritability of ascites and the ratio of right to total ventricle weight in broiler breeder male lines. Poult. Sci. 74:1237–1241. Mitchell, M. A. 1997. Ascites syndrome: A physiological and biochemical perspective. World’s Poult. Sci. 53:61–64. NRC. 1994. Nutrient Requirements of Poultry. 9th rev. ed. Natl. Acad. Press, Washington, DC. Owen, R. L., R. F. Wideman, A. L. Hattel, and B. S. Cowen. 1990. Use of a hypobaric chamber as a model system for investigating ascites in broilers. Avian Dis. 34:754–758. Pakdel, A., P. Bijma, B. J. Ducro, and H. Bovenhuis. 2005. Selection strategies for body weight and reduced ascites susceptibility in broilers. Poult. Sci. 84:528–535. Pakdel, A., J. A. M. van Arendonk, A. L. J. Vereijken, and H. Bovenhuis. 2002. Direct and maternal genetic effects for ascites-related traits in broilers. Poult. Sci. 81:1273–1279. Riddell, C. 1991. Developmental, metabolic, and miscellaneous disorders. Pages 839–841 in Diseases of Poultry. 9th ed. B. W. Calnek, H. J. Barnes, C. W. Beard, W. M. Reid, and H. W. Yoder Jr., ed. Iowa State Univ. Press, Ames. Roush, W. B., and R. F. Wideman Jr. 2000. Evaluation of broiler growth velocity and acceleration in relation to pulmonary hypertension syndrome. Poult. Sci. 79:180–191. SAS Institute. 1988. SAS/STAT User’s Guide. 1988 Edition. SAS Inst. Inc., Cary, NC. Siegel, P. B. 1999. Utilizing resources for broilers. World Poult. 15:23–24. Shlosberg, A., M. Bellaiche, V. Hanji, A. Nyska, A. Lublin, M. Shemesh, L. Shore, S. Perk, and E. Berman. 1996. The effect of acetylsalicylic acid and cold stress on the susceptibility of broiler to the ascites syndrome. Avian Pathol. 25:581–590. Shlosberg, A., I. Zadikov, U. Bendheim, V. Handji, and E. Berman. 1992. The effects of poor ventilation, low temperatures, type of feed and sex on bird on the development of ascites in broilers. Physiopathological factors. Avian Pathol. 21:369–382. Vanhooser, S. L., A. Becker, and R. G. Teeter. 1995. Bronchodilator, oxygen level, and temperature effects on ascites incidence in broiler chickens. Poult. Sci. 74:1586–1590. Verstegen, M. W. A., A. M. Henken, W. Van der Hel, and M. T. Frankenhuis. 1989. Pages 65–69 in Proc. 7th Int. Conf. Prod. Dis. Farm Anim., Cornell Univ., Ithaca, NY. Wideman, R. F. 2000. Cardio-pulmonary hemodynamics and ascites in broiler chickens. Avian Poult. Biol. Rev. 11:21–43. Wideman, R. F., Jr., and W. G. Bottje. 1993. Current understanding of the ascites syndrome and future research directions. Pages 1–20 in Nutr. Tech. Symp. Proc. Novus Int. Inc., St. Louis, MO. Wideman, R. F., and G. F. Erf. 2002. Intravenous micro-particle injection and pulmonary hypertension in broiler chickens: Cardio-pulmonary hemodynamic responses. Poult. Sci. 81:877–886. Wideman, R. F., Jr., and H. French. 2000. Ascites resistance of progeny from broiler breeders selected for two generations using chronic unilateral pulmonary artery occlusion. Poult. Sci. 79:396–401. Wideman, R. F., Jr., and Y. K. Kirby. 1995. A pulmonary artery clamp model for inducing pulmonary hypertension syndrome (ascites) in broilers. Poult. Sci. 74:805–812. Wideman, R. F., Jr., Y. K. Kirby, R. L. Owen, and H. French. 1997. Chronic unilateral occlusion of an extrapulmonary primary bronchus induces pulmonary hypertension syndrome (ascites) in male and female broilers. Poult. Sci. 76:400–404. Witzel, D. A., W. E. Huff, L. F. Kubena, R. B. Harvey, and M. H. Elissalde. 1990. Ascites in growing broilers: A research model. Poult. Sci. 69:741–745.
Downloaded from http://ps.oxfordjournals.org/ at National Chung Hsing University Library on April 11, 2014
Anthony, N. B., J. M. Balog, J. D. Hughes Jr., L. Stamps, M. A. Cooper, B. D. Kidd, X. Liu, G. R. Huff, W. E. Huff, and N. C. Rath. 2001. Genetic selection of broiler lines that differ in their ascites susceptibility 1. Selection under hypobaric conditions. Pages 327–328 in Proc. 13th Eur. Symp. Poult. Nutr., Blankenberge, Belgium. Balog, J. M. 2003. Ascites syndrome (pulmonary hypertension syndrome) in broiler chickens: Are we seeing the light at the end of the tunnel? Avian Poult. Biol. Rev. 14:99–126. Balog, J. M., N. B. Anthony, M. A. Cooper, B. D. Kidd, G. R. Huff, W. E. Huff, and N. C. Rath. 2000. Ascites syndrome and related pathologies in feed restricted broilers raised in a hypobaric chamber. Poult. Sci. 79:318–323. Beker, A., S. L. Vanhooser, J. H. Swartzlander, and R. G. Teeter. 2003. Graded atmospheric oxygen level effects on performance and ascites incidence in broilers. Poult. Sci. 82:1550– 1553. Boldman, K. L., A. Kriese, L. D. Van Vleck, and S. D. Kachman. 1993. A manual for the use of MTDFREML—A set of programs to obtain estimates of variances and covariances. ARS, USDA, Washington, DC. Buys, N., C. W. Scheele, C. Kwakernaak, and E. Decuypere. 1999. Performance and physiological variables in broiler chicken lines differing in susceptibility to the ascites syndrome: 2. Effect of ambient temperature on partial efficiencies of protein and fat retention and plasma hormone concentrations. Br. Poult. Sci. 40:140–144. Christensen, V. L., D. O. Noble, and K. E. Nestor. 2000. Influence of selection for increased body weight, egg production, and shank width on the length of the incubation period of turkeys. Poult. Sci. 79:613–618. De Greef, K. H., L. L. G. Janss, A. L. J. Vereijken, R. Pit, and C. L. M. Gerritsen. 2001. Disease-induced variability of genetic correlations: Ascites in broilers as a case study. J. Anim. Sci. 79:1723–1733. Dunnington, E. A., and P. B. Siegel. 1998. Resource allocations: Growth and immune responses. Pages 95–98 in Proc. 10th Eur. Poult. Conf. Falconer, D. S., and T. F. C. Mackay. 1996. Introduction to Quantitative Genetics. Pearson Educ. Ltd., Essex, UK. Julian, R. J. 1993. Ascites in poultry. Avian Pathol. 22:419–454. Julian, R. J. 1998. Rapid growth problems: Ascites and skeletal deformities in broilers. Poult. Sci. 77:1773–1780. Julian, R. J. 2000. Physiological, management and environmental triggers of the ascites syndrome: A review. Avian Pathol. 29:519–527. Julian, R. J., G. W. Friars, H. French, and M. Quinton. 1986. The relationship of right ventricular hypertrophy, right ventricular failure, and ascites to weight gain in broiler and roaster chickens. Avian Dis. 31:130–135. Julian, R. J., I. McMillan, and M. Quinton. 1989. The effect of cold and dietary energy on right ventricular hypertrophy, right ventricular failure and ascites in meat-type chickens. Avian Pathol. 18:675–684. Julian, R. J., and B. Wilson. 1992. Pen oxygen concentration and pulmonary hypertension induced right ventricular failure and ascites in meat type chickens at low altitude. Avian Dis. 36:733–735. Lister, S. 1997. Broiler ascites: A veterinary viewpoint. World’s Poult. Sci. 53:65–67.
2529