- Email: [email protected]
Divergent Selection for Incidence of Degenerative Myopathy of the Musculus supracoracoideus of Meat-Type Chickens1
Divergent Selection for Incidence of Degenerative Myopathy of the Musculus supracoracoideus of Meat-Type Chickens1 K. G. HOLLANDS,2 A. A. GRUNDER, 3 and J. S. GAVORA Animal Research Centre, Agriculture Canada, Ottawa, Ontario, Canada K1A 0C6 (Received for publication May 10, 1985)
1986 Poultry Science 65:417-425
INTRODUCTION C o m m o n l y called green muscle disease, degenerative m y o p a t h y of t h e Musculus supracoracoideus (DMS) affects only t h e deep pectoral muscle which elevates t h e wing. T h e disease was first described in t u r k e y s b y Dickinson et al. ( 1 9 6 8 ) and in m e a t - t y p e chickens b y Harper et al. ( 1 9 7 1 ) . It affects mainly adult birds b u t has been observed m o r e recently in 7-week-old broiler chickens (Richardson et al., 1980). T h e early stages of DMS are characterized b y e d e m a and h e m o r r h a g e of t h e M. supracoracoideus {Pectoralis minor). Later, o n e or m o r e encapsulated areas of green fibrous tissue will appear followed b y varying degrees of a t r o p h y of t h e muscle and partial replacement by fibrous and f a t t y tissue. Siller et al. ( 1 9 7 9 b ) have been able t o produce t h e disease experimentally b y inducing birds t o flap their wings until exhausted. T h e y have presented evidence suggesting t h a t muscle
fascia prevents muscle enlargement in s o m e of t h e exercised birds with a resultant lack of b l o o d supply and ensuing lesion d e v e l o p m e n t (Siller et al, 1979a). In t u r k e y s , Hollands et al. ( 1 9 8 1 ) have s h o w n t h a t plasma creatine kinase (CK) levels 4, 5, and 6 days postexercise are significantly higher in birds t h a t eventually b e c o m e affected b y DMS c o m p a r e d with t h o s e t h a t are not. Harper et al. ( 1 9 8 3 ) has used induced wing exercise in t u r k e y s t o identify susceptible birds at prebreeding ages. No studies have been r e p o r t e d w h e r e changes in incidence of DMS in chickens have been a t t e m p t e d . Harper et al. ( 1 9 7 5 ) d e m o n strated t h a t incidence of DMS in t u r k e y s could be increased t h r o u g h selection and t h a t t h e disease is genetically controlled b y m u l t i p l e genes. T h e p u r p o s e of t h e present s t u d y was t o d e t e r m i n e if incidence of DMS in m e a t - t y p e chickens could be changed t h r o u g h selection and t o estimate heritability of DMS incidence, and genetic and p h e n o t y p i c correlations of DMS w i t h economically i m p o r t a n t traits.
'Contribution number 1323 from the Animal Research Centre. 2 Deceased 1983. 3 To whom correspondence should be addressed. 417
MATERIALS AND METHODS Stocks and Selection Procedures. Parent stocks of m e a t - t y p e chickens from t w o com-
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
ABSTRACT The response to three generations of selection for incidence of degenerative myopathy of the Musculus supracoracoideus (DMS) in meat-type chickens was studied. From each of two genetic bases named E and F, which were derived from two commercial broiler stocks, a high incidence, low incidence, and a nonselected control strain were initiated. Selection of breeders to produce Generations 1 and 2 was based on the incidence of DMS in their parents after induced wing exercise: mass selection for plasma creatine kinase levels after induced wing exercise was used to produce Generation 3. Little genetic progress in selection occurred in Generations 1 and 2. There was a response in Generation 3, especially in Base E selected for low incidence. Heritability of incidence of DMS was estimated to be .48. Body weights at 8 weeks of age, 18 weeks of age, and breast angle at 18 weeks of age of Generation 3 were greater (P<.05) for affected than nonaffected birds and positively correlated phenotypically with incidence of DMS. Genetic correlations were extremely variable. The higher incidence of DMS in birds with desirable traits suggests that the breeder must use a multitrait selection program to reduce incidence of DMS as well as improve production traits. (Key words: genetic selection, degenerative myopathy, chickens, body size, breast angle)
418
HOLLANDS ET AL.
To facilitate the selection of Generation 1 breeders for producing Generation 2, induced wing exercise was practiced on the Generation 0 breeders. They were killed a week later at 62 weeks of age and classified for DMS. To produce the C strains of Generation 2, an effort was made to choose one male breeder at random from each sire family and similarly choose one female breeder from each dam family. In the H and L strains, breeders were selected from families whose parents had the desired diagnosis of DMS, i.e., the breeders in the L strains were from parents free of DMS, while for the H strains, the breeders were from families where one and preferably both parents
bA
c o
I II 1
O -o rt
'" ° E « S .5
en E «s 2 T3 2 C ^ C (U
&
Si !
« OS
fjffji PH
K
[X. [Li
K
t-
i-t
00
0\
O
T-H
N N
+
T»-
PH
^ N m -f
H in
|
|
+
t-- \0 i n 00 rH so (M y\
TH
+ + + + +
H
N
fft H
VO rt
fft \o f i rn irt ro
« E s
f~ —1 00 en | N
V)
5 + t1 1 + + -~. > IN (N O* ~ ~ irv o o
7-t
"*3 S
VI N
00
H l f t N
+ 1 I + + +
-a
s
o
UJ CO
< ea •3
X O J X O J W W W H, H. b
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
mercial companies were designated genetic Bases E and F. The hens were force molted after the first year of egg production and after starting the second egg production cycle, they were mated within each genetic base to produce pedigreed Generation 0. The parents of Generation 0, designated as Generation-1, were wing exercised (Siller et al, 1979b) at 94 weeks of age. Each bird was exercised by grasping its hocks and tilting it backwards until it was off balance and started to flap its wings. The flapping was allowed to continue until wingbeats were infrequent or ceased. An hour later, the process was repeated. A week later, the birds were killed and classified for DMS by necropsy. On the basis of DMS incidence in the parents, an unselected control strain (C), a high DMS incidence strain (H), and a low DMS incidence strain (L) of breeders were derived within each base as follows. Randomly selected males and females of Generation 0 were mated to produce the C strains; Generation 0 progeny of DMS positive parents were used, as far as possible, to produce the H strains, and the Generation 0 progeny of DMS negative parents were mated to produce the L strains of Generation 1. While breeders of Generation 0 were wing exercised at 61 weeks of age and killed a week later, nonbreeders were exercised at 54 weeks of age and killed 3 weeks later. The numbers of birds mated in each strain derivation and in subsequent generations are recorded in Table 1. Hereafter, the individual strains will be referred to as EC, EH, and EL for those derived from the E base and analogically, FC, FH, and FL for those originating from the F base. Once selected, all birds in this and subsequent generations were mated at random, avoiding full-sib matings, and all matings were by artificial insemination.
50 48 50 55 49 46
124* 189 128** 81 149 114
EH EC EL FH FC FL
EH EC EL FH FC FL
105* 154 124 113 150 92
118 163 101 111** 179 107
129 215 139** 100* 193 181
6 15 6 5 6 2
2 8 4 0 4 0
32 21 7 10 9 7
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
67 59 26 54 41 38
No
No
No No No No
18 10 11 15 6 4
20 10 6 6 6 1
14 31 16 14 29 16
18 32 18 17 32 17 CK + 7 +2.8 + 2 -.08 + 3 -.7 0 +1.3 -12 -.05 0 -1.2
+5 +1 0 +7 -14 -8 ( +43/+40) ( -2/+24) (-12/ - 8 ) (+53/+50) ( +2/ +2) (-51/-46)
31 60 29 31 62 30
36 64 36 34 64 34
Birds in Generation 3 were killed at 21 weeks of age just prior to housing. Only progeny of parents with recorded DM
*P<.05 and **P<.01 are probabilities of Chi-square values for homogeneity of incidence of DMS for Strain EH or ELra
2
1 Selection of breeders of Generations 0 and 1 was based on degenerative myopathy of the Musculus spracoracoideu Generations—1 and 0, respectively. The figures in brackets show the S of the breeders parents: sires/dams. Selection of b the trait plasma creatine kinase (CK), expressed as international units per milliliter, and thus S for DMS incidence as well a
47 53 51 55 52 48
EH EC EL FH FC FL
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
420
HOLLANDS ET AL.
the sires' parents was calculated as the arcsin y/% DMS of sires' mean minus the arcsin \/% DMS of the mean of the population from which the sires were selected. The S of the dams' parents was calculated similarly. Selection differentials of Generation 2 breeders for plasma CK were calculated directly, and those for DMS were calculated retrospectively after the postmortem examination. Heritability of incidence of DMS was estimated in Generation 3 only. Sexes were considered separately and only strains within sex were considered if % DMS were greater than 10 (Table 1). Heritability on the binomial scale was estimated by the Chi-square method using sire family data (Robertson and Lerner, 1949) and converted to a normal scale (Dempster and Lerner, 1950). These same data, after scoring birds 0 for negative and 1 for positive, were subjected to nested analysis of variance, and heritability was estimated from sire variances. The same DMS data and data on production traits were subjected to nested analyses of covariance, and genetic and phenotypic correlations were estimated for strains combined within sex (Table 2). Differences in DMS incidence in Generation 3 among lines, within base and sex, were assessed by Chi-squares of homogeneity (Steel and Torrie, 1960), and body weight and breast angle were also analyzed by analysis of variance considering the fixed effects of DMS, sex, genetic base and strain, and their two-way interactions (Steel and Torrie, 1960).
RESULTS
Selection Practiced. Selection for and against DMS had to be accomplished in an indirect manner because birds could only be diagnosed after necropsy. Breeders of Generations 0 and 1 were selected on the basis of incidence of DMS in their parents after forced wing exercise. Selection differentials, expressed as arcsin y/% DMS, were relatively large for breeders of Generation 0 (—8.1 to +28.5) but generally smaller for Generation 1 (—14.2 to +6.6) (Table 1). In some instances, the S was not in the desired direction, e.g., Strain EL female breeders and Strain FL breeders of both sexes in Generation 0 were, in fact, selected toward a higher incidence of DMS, although the intention was to select in the opposite direction. As mentioned earlier, the selection of breeders was based on DMS incidence in their parents (grand-
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
were DMS positive. Birds of Generation 1 were not wing exercised before being killed at 60 weeks of age, because the authors wished to observe incidence of DMS under normal management practices. To produce Generation 3, birds of Generation 2 were exercised at 18 weeks of age, bled 5 days later, and plasmas were assayed for CK (Hollands et al, 1981). On the basis of plasma CK levels, and within base and strain, matings were made such that birds with high plasma CK were used to reproduce the H strains, and those with low plasma CK were used to reproduce the L strains. Within families of the control strains, breeders were chosen at random as described above. After reproducing Generation 3, Generation 2 birds were killed at 47 weeks of age. Birds of Generation 3 were wing exercised at 19 or 20 weeks of age and killed 16 and 9 days postexercise, respectively, to study plasma CK as an indicator of DMS (Hollands et al, 1983). Husbandry. All stocks were brooded and reared on the floor with litter but were housed individually in cages at 126 to 138 days of age, except that Generation —1 was not caged until after its first year of production, just after being molted. Mating was by artificial insemination. Diagnosis for Degenerative Myopathy of the Museulus supracoracoideus. The experimental birds killed or those that died during the study were examined postmortem for lesions related to DMS. Birds with myopathy on either or both sides were considered positive for DMS. Plasma Creatine Kinase Analysis. Blood was collected from the brachial vein of Generation 2 birds into heparinized tubes 5 days after exercise at 18 weeks of age. Plasma was stored at —20 C until assayed for CK by method No. SE4-0017FG5 of Technicon Instruments Corporation, Tarrytown, NY. Selection Differential, Heritability, Correlation Estimation, and Statistical Analyses. Selection differential (S) for percent DMS of breeders, selected on the basis of DMS incidence in their parents (grandparents of the birds produced), were calculated retrospectively as the arcsin y/% DMS of their mean minus the arcsin ->/% DMS of the mean of the population from which they were selected. The S of the parents of breeders of Generations 0 and 1, i.e., grandparents of Generations 1 and 2, respectively, were also calculated because only data on the breeders' parents were actually available and used at the time of selection. The S of
DEGENERATIVE MYOPATHY IN CHICKENS
421
TABLE 2. Estimates of heritability (h2)1 with standard errors (SE) of incidence of degenerative myopathy of the Musculus supracoracoideus (DMS) and pheno typic correlations with production traits of Generation 3 Phenotypic correlations, binomial (normal) 3
Chi-square
Straur
Sex
h2
SE
Normal h2
EH EC E(H+C)
Male
.42 .19
.25 .17
.71 .39
EH EC EL E(H+C+L)
Female
FH FC FL F(H+C+L)
Female
Bino-
.54 .26 .35
.49 .11 .08
.26 .16 .22
.29 .15 .12
.90 .42 .64
.78 .17 .14
Analysis of variance h2
SE
.77 .17 .45
.36 .25 .20
.82 .34 .36 .40
.40 .27 .36 .20
.82 .22 .29 .36
.38 .25 .18 .15
Body weight 8 wk 18 wk
angle
.10 (.14)
.29** (.39)
.28** (.38)
.17 (.21)
.29** (.36)
.36** (.45)
.07 (.09)
.22** (.28)
.27** (.34)
1 The Chi-square method was that of Robertson and Lerner (1949) and Dempster and Lerner (1950), and the analysis of variance was as described by Becker (1975) with SE estimated by the method of Dickerson (1960). 2 The H, C, and L represent strains selected for high, with no selection, and selected for low incidence of DMS, respectively, in each of Bases E and F. 'Conversion of binomial correlations to the normal scale was done by multiplying the binomial correlation by P(l — P)/Z where Z = ordinate of the point of truncation at P = incidence of DMS (A. J. McAllister, personal communication).
**P<.01.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
parents of the progeny produced). In terms of plotted in Figure 1, within strain and sex, for breeders' parents, the selection was always in Generations 1 to 3. Virtually no response to the desired direction except in the case of selection was obtained for Generations 1 and 2. Strain EC, Generation 1, where S for breeders' For Generation 3, there was an apparent response in the downward direction for Strain EL dams was +24 to 25 rather than the desired 0. As mentioned earlier, the selection of males and females. Generation 2 breeders was based on the level of Heritability of Degenerative Myopathy of the CK, and this the S values for DMS shown in Musculus supracoracoideus, Heritabilities were Table 1 were calculated after reproduction calculated from data of Generation 3 because and necropsy of the breeders. Unfortunately, it had the highest incidence of DMS among this indirect selection for DMS was also inac- the three selected generations (Table 1, Fig. 1). curate mainly in the downward direction as can Heritabilities (Table 2) estimated by the Chibe seen from Table 1. Nevertheless, the S values square method for the normal scale had an for level of CK were in the desired direction average of .52 and ranged from .14 to .90. (Table 1). Average heritability estimated by analysis of Selection Response. Incidence of DMS variance was .45 and ranged from .22 to .82. tended to be higher for females than males, The mean of the two averages is .48 based especially in Generation 3 (Table 1) where Chi- on data from one generation. square values for sex comparisons within strain Production Traits and Incidence of Deranged from 17 to 60 (P<.01). Hence, inci- generative Myopathy of the Musculus supradence of DMS was compared in Table 1 and coracoideus. These traits, namely, 8-week-
422
HOLLANDS ET AL.
angle of Generation 3 are summarized in Table 3. Affected birds had significantly (P<.05) greater body weights and breast angle than nonaffected birds. Differences between sexes, genetic bases, and among strains did not always reflect the incidences of DMS shown in Table 1. There were significant (P<.05) two-way interactions for 8-week body weight (base X strain), 18-week body weight (necropsy class X sex, base
x strain, and sex X strain), and 18-week breast angle (autopsy class X sex). However, the strain selected for low incidence of DMS tended to have the lowest DMS incidence (Table 1) and had the lowest 18-week body weight and breast angle (Table 3). Phenotypic and genetic correlations between the body traits and incidence of DMS were estimated in the same populations from which
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
GENERATION
FIG. 1. Response to selection for incidence of degenerative myopathy of the Musculus supracoracoideus (DMS) plotted as deviations from incidence recorded for the control strains after percentages had been transformed by arcsin yjpercentage. The H, C, and L represent strains selected for high, with no selection, and selected for low incidence of DMS, respectively, in each of Bases E and F.
423
DEGENERATIVE MYOPATHY IN CHICKENS
DISCUSSION
There were two different methods of selection for incidence of DMS, namely, pedigree selection on DMS incidence after exercise for producing Generations 1 and 2, and mass selection for the related trait plasma CK for Generation 3. No direct selection for DMS could be practiced because necropsy was necessary for diagnosis. Response to selection for DMS was low or negligible, probably due to the erratic and low selection differentials caused by the indirect selection. When the selection was expressed in terms of DMS incidence of the parents themselves, which was determined only retrospectively after the parents were reproduced, killed, and necropsied, the parental S sometimes differed dramatically from those for the breeders' parents, not only in magnitude
TABLE 3. Least squares means and significance of differences in body weight and breast angle after 3 generations of selection for degenerative myopathy of the Musculus supracoracoideus (DMS)
Source of variation
No.'
18-wk Body weight
8-wk Body weight N"g>
~
18-wk Breast angle
o
*
***
558 1131
220.5 217.1
334.4 312.4
Sex Male Female
***
** *
***
749 940
243.6 193.9
361.1 285.6
92.5 97.8
Base E F
***
***
897 792
219.8 217.7
329.5 317.3
94.3 96.1
Strain High Control Low
*
*
* **
418 725 546
217.5 220.6 218.2
324.1 326.7 319.3
96.3 95.6 93.6
Necropsy class Affected Nonaffected
*** 97.4 92.9
1 Only birds with a recorded DMS status when killed at 21 weeks of age and measured for all three production traits are included.
*P<.05. ***P<.001.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
but also in sign. For example in Strain FL, in Generation 0, breeders' S values were +2 and +6 for sires and dams, respectively, but were —51/ —46 for sires' and dams' parents (Table 1). The greatest selection response (Fig. 1) was observed in Generation 3, indicating that indirect selection on the basis of plasma CK response to exercise could be effective. Selection in the downward direction in the L strains was most successful in the EL strain, the E strains having a higher DMS incidence in general, and thus a greater potential for improvement, than the F strain. Results of selection in Generation 3 indicated that selection for a continuous trait, plasma CK, related to DMS incidence and measured in chickens to be selected, may be more effective than selection for the threshold trait of interest, DMS, measured only in parents of the chickens to be selected. In direct selection on CK levels, the S in terms of DMS incidence may not be an adequate description of the selection practiced, because the level of CK may have been an indication of DMS susceptibility even though the DMS itself may not have been expressed in
heritability estimates were reported above. Phenotypic correlations, shown in Table 2, were all positive. The comparable genetic correlations (not shown) were much more variable in that they ranged from —1.26 to .54 with about half of them being positive.
424
HOLLANDS ET AL. ACKNOWLEDGMENTS
The authors wish to thank P. Griffin for skillful technical help and C. Williams for discussion of statistical procedures. REFERENCES Becker, W. A., 1975. Manual of Quantitative Genetics. Univ. Press, Washington State Univ., Pullman, WA. Dempster, E. R., and I. M. Lemer, 1950. Heritability of threshold characters. Genetics 35:212—236. Dickerson, G. E., 1960. Techniques for research in quantitative animal genetics. Pages 57—96 in Techniques and Procedures in Animal Production Research. Am. Soc. of Anim. Prod., Champaign, IL. Dickinson, E. M., J. O. Stevens, and D. H. Heifer, 1968. A degenerative myopathy in turkeys. Page 6 in Proc. 17th West. Poult. Dis. Conf., Univ. Calif., Davis, CA. Falconer, D. S., 1976. Pages 3 0 1 - 3 1 1 in Introduction to Quantitative Genetics. The Ronald Press Company, New York, NY. Grunder, A. A., K. G. Hollands, J. S. Gavora, J. R. Chambers, and N.A.G. Cave, 1984. Degenerative myopathy of the Musculus supracoracoideus and production traits in strains of meat-type chickens. Poultry Sci. 63:781-785. Harper, J. A., P. E. Bernier, D. H. Heifer, and J. A. Schmitz, 1975. Degenerative myopathy of the deep pectoral muscle in the turkey. J. Hered. 66: 362-366. Harper, J. A., P. E. Bernier, and L. L. ThompsonCowley, 1983. Early expression of hereditary deep pectoral myopathy in turkeys due to forced wing exercise. Poultry Sci. 62:2303-2308. Harper, J. A., D. H. Heifer, and E. M. Dickinson, 1971. Hereditary myopathy in turkeys. Page 76 in Proc. 20th West. Poult. Dis. Conf., Univ. of Calif., Davis, CA. Hollands, K. G., A. A. Grunder, and J. S. Gavora, 1983. Plasma creatine kinase as an indicator of degenerative myopathy of the M. supracoracoideus (DMS) in exercised chickens. Poultry Sci. 62:1435. (Abstr.) Hollands, K. G., A. A. Grunder, C. J. Williams, J. S. Gavora, J. R. Chambers, and N.A.G. Cave, 1981. Degenerative myopathy in meat-type poultry: its effect on production traits in chickens and its identification in live turkeys. Pages 337—344 in Quality of Meat. Proc. 5th Eur. Symp., Apeldoorn, The Netherlands. Richardson, J. A., J. Burgener, R. W. Winterfield, and A. S. Dhillon, 1980. Deep pectoral myopathy in seven-week-old broiler chickens. Avian Dis. 24: 1054-1059. Robertson, A., and I. M. Lemer, 1949. The heritability of all-or-none traits: viability of poultry. Genetics 34:395-411. Siller, W. G„ L. Martindale, and P.A.L. Wight, 1979a. The prevention of experimental deep pectoral myopathy of the fowl by fasciotomy. Avian Pathol. 8:301-307. Siller, W. G., P.A.L. Wight, and L. Martindale, 1979b.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
the bird. The CK level, a continuous variable, could be regarded as an expression of the underlying continuous character, similar to that in the general concept of threshold characters (Falconer, 1976). Incidence of DMS can then be viewed as the all or none trait expressed only when the continuous underlying character exceeds a threshold. The heritability estimates for incidence of DMS were quite variable (Table 2). The Chisquare method of estimating heritability from binomial data was based on half-sib relationships, and thus conversion to the normal scale is reasonable, according to Van Vleck (1972), when the proportion of affected individuals is not too small and normal heritability is not too large. The heritability estimate of .48 is at least medium, and based on this heritability, and assuming birds affected by DMS can be identified accurately when alive, a breeder can expect progress in selecting against DMS. Induced wing exercise and measurement of plasma CK (Hollands et al, 1983) should enable him to select breeders more accurately than if he uses family selection. In any selection program for improving meat-type chickens, growth rate and conformation as well as other traits are most important. Data of Tables 2 and 3 suggest that the desirable traits such as fast growth rate and large breast angle are greater in DMS positive than DMS negative birds. These observations are consistent with those of Hollands et al. (1981) and Grunder et al. (1984) who presented evidence that faster-growing birds with good conformation were more susceptible to DMS than slowergrowing birds with bad conformation. Also in agreement were the results of Siller et al. (1979b) who were able to cause DMS in turkeys and broiler chicken stocks but not in slower growing, narrow breasted, egg-type chickens by forced wing exercise. The phenotypic correlations of Table 2 also support these observations. Selection for body weight and conformation alone would lead to DMS susceptibility (predisposition) as apparently happened in commercial broiler chicken stocks (Grunder et al, 1984). Selection against DMS alone may lead to reduction in body weight and conformation. Therefore, a multitrait approach, combining selection for rapid growth and body conformation with selection against DMS susceptibility, would be best for improving meat type stocks.
DEGENERATIVE MYOPATHY IN CHICKENS
Exercise-induced deep pectoral myopathy in broiler fowls and turkeys. Vet. Sci. Commun. 2: 331-336. Steel, R.G.D., and J. H. Torrie, 1960. Principles and
425
Procedures of Statistics. McGraw-Hill, New York, NY. Van Vleck, L. D., 1972. Estimation of heritability of threshold characters. J. Dairy Sci. 55:218-225.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
1986 Poultry Science 65:417-425
INTRODUCTION C o m m o n l y called green muscle disease, degenerative m y o p a t h y of t h e Musculus supracoracoideus (DMS) affects only t h e deep pectoral muscle which elevates t h e wing. T h e disease was first described in t u r k e y s b y Dickinson et al. ( 1 9 6 8 ) and in m e a t - t y p e chickens b y Harper et al. ( 1 9 7 1 ) . It affects mainly adult birds b u t has been observed m o r e recently in 7-week-old broiler chickens (Richardson et al., 1980). T h e early stages of DMS are characterized b y e d e m a and h e m o r r h a g e of t h e M. supracoracoideus {Pectoralis minor). Later, o n e or m o r e encapsulated areas of green fibrous tissue will appear followed b y varying degrees of a t r o p h y of t h e muscle and partial replacement by fibrous and f a t t y tissue. Siller et al. ( 1 9 7 9 b ) have been able t o produce t h e disease experimentally b y inducing birds t o flap their wings until exhausted. T h e y have presented evidence suggesting t h a t muscle
fascia prevents muscle enlargement in s o m e of t h e exercised birds with a resultant lack of b l o o d supply and ensuing lesion d e v e l o p m e n t (Siller et al, 1979a). In t u r k e y s , Hollands et al. ( 1 9 8 1 ) have s h o w n t h a t plasma creatine kinase (CK) levels 4, 5, and 6 days postexercise are significantly higher in birds t h a t eventually b e c o m e affected b y DMS c o m p a r e d with t h o s e t h a t are not. Harper et al. ( 1 9 8 3 ) has used induced wing exercise in t u r k e y s t o identify susceptible birds at prebreeding ages. No studies have been r e p o r t e d w h e r e changes in incidence of DMS in chickens have been a t t e m p t e d . Harper et al. ( 1 9 7 5 ) d e m o n strated t h a t incidence of DMS in t u r k e y s could be increased t h r o u g h selection and t h a t t h e disease is genetically controlled b y m u l t i p l e genes. T h e p u r p o s e of t h e present s t u d y was t o d e t e r m i n e if incidence of DMS in m e a t - t y p e chickens could be changed t h r o u g h selection and t o estimate heritability of DMS incidence, and genetic and p h e n o t y p i c correlations of DMS w i t h economically i m p o r t a n t traits.
'Contribution number 1323 from the Animal Research Centre. 2 Deceased 1983. 3 To whom correspondence should be addressed. 417
MATERIALS AND METHODS Stocks and Selection Procedures. Parent stocks of m e a t - t y p e chickens from t w o com-
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
ABSTRACT The response to three generations of selection for incidence of degenerative myopathy of the Musculus supracoracoideus (DMS) in meat-type chickens was studied. From each of two genetic bases named E and F, which were derived from two commercial broiler stocks, a high incidence, low incidence, and a nonselected control strain were initiated. Selection of breeders to produce Generations 1 and 2 was based on the incidence of DMS in their parents after induced wing exercise: mass selection for plasma creatine kinase levels after induced wing exercise was used to produce Generation 3. Little genetic progress in selection occurred in Generations 1 and 2. There was a response in Generation 3, especially in Base E selected for low incidence. Heritability of incidence of DMS was estimated to be .48. Body weights at 8 weeks of age, 18 weeks of age, and breast angle at 18 weeks of age of Generation 3 were greater (P<.05) for affected than nonaffected birds and positively correlated phenotypically with incidence of DMS. Genetic correlations were extremely variable. The higher incidence of DMS in birds with desirable traits suggests that the breeder must use a multitrait selection program to reduce incidence of DMS as well as improve production traits. (Key words: genetic selection, degenerative myopathy, chickens, body size, breast angle)
418
HOLLANDS ET AL.
To facilitate the selection of Generation 1 breeders for producing Generation 2, induced wing exercise was practiced on the Generation 0 breeders. They were killed a week later at 62 weeks of age and classified for DMS. To produce the C strains of Generation 2, an effort was made to choose one male breeder at random from each sire family and similarly choose one female breeder from each dam family. In the H and L strains, breeders were selected from families whose parents had the desired diagnosis of DMS, i.e., the breeders in the L strains were from parents free of DMS, while for the H strains, the breeders were from families where one and preferably both parents
bA
c o
I II 1
O -o rt
'" ° E « S .5
en E «s 2 T3 2 C ^ C (U
&
Si !
« OS
fjffji PH
K
[X. [Li
K
t-
i-t
00
0\
O
T-H
N N
+
T»-
PH
^ N m -f
H in
|
|
+
t-- \0 i n 00 rH so (M y\
TH
+ + + + +
H
N
fft H
VO rt
fft \o f i rn irt ro
« E s
f~ —1 00 en | N
V)
5 + t1 1 + + -~. > IN (N O* ~ ~ irv o o
7-t
"*3 S
VI N
00
H l f t N
+ 1 I + + +
-a
s
o
UJ CO
< ea •3
X O J X O J W W W H, H. b
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
mercial companies were designated genetic Bases E and F. The hens were force molted after the first year of egg production and after starting the second egg production cycle, they were mated within each genetic base to produce pedigreed Generation 0. The parents of Generation 0, designated as Generation-1, were wing exercised (Siller et al, 1979b) at 94 weeks of age. Each bird was exercised by grasping its hocks and tilting it backwards until it was off balance and started to flap its wings. The flapping was allowed to continue until wingbeats were infrequent or ceased. An hour later, the process was repeated. A week later, the birds were killed and classified for DMS by necropsy. On the basis of DMS incidence in the parents, an unselected control strain (C), a high DMS incidence strain (H), and a low DMS incidence strain (L) of breeders were derived within each base as follows. Randomly selected males and females of Generation 0 were mated to produce the C strains; Generation 0 progeny of DMS positive parents were used, as far as possible, to produce the H strains, and the Generation 0 progeny of DMS negative parents were mated to produce the L strains of Generation 1. While breeders of Generation 0 were wing exercised at 61 weeks of age and killed a week later, nonbreeders were exercised at 54 weeks of age and killed 3 weeks later. The numbers of birds mated in each strain derivation and in subsequent generations are recorded in Table 1. Hereafter, the individual strains will be referred to as EC, EH, and EL for those derived from the E base and analogically, FC, FH, and FL for those originating from the F base. Once selected, all birds in this and subsequent generations were mated at random, avoiding full-sib matings, and all matings were by artificial insemination.
50 48 50 55 49 46
124* 189 128** 81 149 114
EH EC EL FH FC FL
EH EC EL FH FC FL
105* 154 124 113 150 92
118 163 101 111** 179 107
129 215 139** 100* 193 181
6 15 6 5 6 2
2 8 4 0 4 0
32 21 7 10 9 7
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
67 59 26 54 41 38
No
No
No No No No
18 10 11 15 6 4
20 10 6 6 6 1
14 31 16 14 29 16
18 32 18 17 32 17 CK + 7 +2.8 + 2 -.08 + 3 -.7 0 +1.3 -12 -.05 0 -1.2
+5 +1 0 +7 -14 -8 ( +43/+40) ( -2/+24) (-12/ - 8 ) (+53/+50) ( +2/ +2) (-51/-46)
31 60 29 31 62 30
36 64 36 34 64 34
Birds in Generation 3 were killed at 21 weeks of age just prior to housing. Only progeny of parents with recorded DM
*P<.05 and **P<.01 are probabilities of Chi-square values for homogeneity of incidence of DMS for Strain EH or ELra
2
1 Selection of breeders of Generations 0 and 1 was based on degenerative myopathy of the Musculus spracoracoideu Generations—1 and 0, respectively. The figures in brackets show the S of the breeders parents: sires/dams. Selection of b the trait plasma creatine kinase (CK), expressed as international units per milliliter, and thus S for DMS incidence as well a
47 53 51 55 52 48
EH EC EL FH FC FL
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
420
HOLLANDS ET AL.
the sires' parents was calculated as the arcsin y/% DMS of sires' mean minus the arcsin \/% DMS of the mean of the population from which the sires were selected. The S of the dams' parents was calculated similarly. Selection differentials of Generation 2 breeders for plasma CK were calculated directly, and those for DMS were calculated retrospectively after the postmortem examination. Heritability of incidence of DMS was estimated in Generation 3 only. Sexes were considered separately and only strains within sex were considered if % DMS were greater than 10 (Table 1). Heritability on the binomial scale was estimated by the Chi-square method using sire family data (Robertson and Lerner, 1949) and converted to a normal scale (Dempster and Lerner, 1950). These same data, after scoring birds 0 for negative and 1 for positive, were subjected to nested analysis of variance, and heritability was estimated from sire variances. The same DMS data and data on production traits were subjected to nested analyses of covariance, and genetic and phenotypic correlations were estimated for strains combined within sex (Table 2). Differences in DMS incidence in Generation 3 among lines, within base and sex, were assessed by Chi-squares of homogeneity (Steel and Torrie, 1960), and body weight and breast angle were also analyzed by analysis of variance considering the fixed effects of DMS, sex, genetic base and strain, and their two-way interactions (Steel and Torrie, 1960).
RESULTS
Selection Practiced. Selection for and against DMS had to be accomplished in an indirect manner because birds could only be diagnosed after necropsy. Breeders of Generations 0 and 1 were selected on the basis of incidence of DMS in their parents after forced wing exercise. Selection differentials, expressed as arcsin y/% DMS, were relatively large for breeders of Generation 0 (—8.1 to +28.5) but generally smaller for Generation 1 (—14.2 to +6.6) (Table 1). In some instances, the S was not in the desired direction, e.g., Strain EL female breeders and Strain FL breeders of both sexes in Generation 0 were, in fact, selected toward a higher incidence of DMS, although the intention was to select in the opposite direction. As mentioned earlier, the selection of breeders was based on DMS incidence in their parents (grand-
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
were DMS positive. Birds of Generation 1 were not wing exercised before being killed at 60 weeks of age, because the authors wished to observe incidence of DMS under normal management practices. To produce Generation 3, birds of Generation 2 were exercised at 18 weeks of age, bled 5 days later, and plasmas were assayed for CK (Hollands et al, 1981). On the basis of plasma CK levels, and within base and strain, matings were made such that birds with high plasma CK were used to reproduce the H strains, and those with low plasma CK were used to reproduce the L strains. Within families of the control strains, breeders were chosen at random as described above. After reproducing Generation 3, Generation 2 birds were killed at 47 weeks of age. Birds of Generation 3 were wing exercised at 19 or 20 weeks of age and killed 16 and 9 days postexercise, respectively, to study plasma CK as an indicator of DMS (Hollands et al, 1983). Husbandry. All stocks were brooded and reared on the floor with litter but were housed individually in cages at 126 to 138 days of age, except that Generation —1 was not caged until after its first year of production, just after being molted. Mating was by artificial insemination. Diagnosis for Degenerative Myopathy of the Museulus supracoracoideus. The experimental birds killed or those that died during the study were examined postmortem for lesions related to DMS. Birds with myopathy on either or both sides were considered positive for DMS. Plasma Creatine Kinase Analysis. Blood was collected from the brachial vein of Generation 2 birds into heparinized tubes 5 days after exercise at 18 weeks of age. Plasma was stored at —20 C until assayed for CK by method No. SE4-0017FG5 of Technicon Instruments Corporation, Tarrytown, NY. Selection Differential, Heritability, Correlation Estimation, and Statistical Analyses. Selection differential (S) for percent DMS of breeders, selected on the basis of DMS incidence in their parents (grandparents of the birds produced), were calculated retrospectively as the arcsin y/% DMS of their mean minus the arcsin ->/% DMS of the mean of the population from which they were selected. The S of the parents of breeders of Generations 0 and 1, i.e., grandparents of Generations 1 and 2, respectively, were also calculated because only data on the breeders' parents were actually available and used at the time of selection. The S of
DEGENERATIVE MYOPATHY IN CHICKENS
421
TABLE 2. Estimates of heritability (h2)1 with standard errors (SE) of incidence of degenerative myopathy of the Musculus supracoracoideus (DMS) and pheno typic correlations with production traits of Generation 3 Phenotypic correlations, binomial (normal) 3
Chi-square
Straur
Sex
h2
SE
Normal h2
EH EC E(H+C)
Male
.42 .19
.25 .17
.71 .39
EH EC EL E(H+C+L)
Female
FH FC FL F(H+C+L)
Female
Bino-
.54 .26 .35
.49 .11 .08
.26 .16 .22
.29 .15 .12
.90 .42 .64
.78 .17 .14
Analysis of variance h2
SE
.77 .17 .45
.36 .25 .20
.82 .34 .36 .40
.40 .27 .36 .20
.82 .22 .29 .36
.38 .25 .18 .15
Body weight 8 wk 18 wk
angle
.10 (.14)
.29** (.39)
.28** (.38)
.17 (.21)
.29** (.36)
.36** (.45)
.07 (.09)
.22** (.28)
.27** (.34)
1 The Chi-square method was that of Robertson and Lerner (1949) and Dempster and Lerner (1950), and the analysis of variance was as described by Becker (1975) with SE estimated by the method of Dickerson (1960). 2 The H, C, and L represent strains selected for high, with no selection, and selected for low incidence of DMS, respectively, in each of Bases E and F. 'Conversion of binomial correlations to the normal scale was done by multiplying the binomial correlation by P(l — P)/Z where Z = ordinate of the point of truncation at P = incidence of DMS (A. J. McAllister, personal communication).
**P<.01.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
parents of the progeny produced). In terms of plotted in Figure 1, within strain and sex, for breeders' parents, the selection was always in Generations 1 to 3. Virtually no response to the desired direction except in the case of selection was obtained for Generations 1 and 2. Strain EC, Generation 1, where S for breeders' For Generation 3, there was an apparent response in the downward direction for Strain EL dams was +24 to 25 rather than the desired 0. As mentioned earlier, the selection of males and females. Generation 2 breeders was based on the level of Heritability of Degenerative Myopathy of the CK, and this the S values for DMS shown in Musculus supracoracoideus, Heritabilities were Table 1 were calculated after reproduction calculated from data of Generation 3 because and necropsy of the breeders. Unfortunately, it had the highest incidence of DMS among this indirect selection for DMS was also inac- the three selected generations (Table 1, Fig. 1). curate mainly in the downward direction as can Heritabilities (Table 2) estimated by the Chibe seen from Table 1. Nevertheless, the S values square method for the normal scale had an for level of CK were in the desired direction average of .52 and ranged from .14 to .90. (Table 1). Average heritability estimated by analysis of Selection Response. Incidence of DMS variance was .45 and ranged from .22 to .82. tended to be higher for females than males, The mean of the two averages is .48 based especially in Generation 3 (Table 1) where Chi- on data from one generation. square values for sex comparisons within strain Production Traits and Incidence of Deranged from 17 to 60 (P<.01). Hence, inci- generative Myopathy of the Musculus supradence of DMS was compared in Table 1 and coracoideus. These traits, namely, 8-week-
422
HOLLANDS ET AL.
angle of Generation 3 are summarized in Table 3. Affected birds had significantly (P<.05) greater body weights and breast angle than nonaffected birds. Differences between sexes, genetic bases, and among strains did not always reflect the incidences of DMS shown in Table 1. There were significant (P<.05) two-way interactions for 8-week body weight (base X strain), 18-week body weight (necropsy class X sex, base
x strain, and sex X strain), and 18-week breast angle (autopsy class X sex). However, the strain selected for low incidence of DMS tended to have the lowest DMS incidence (Table 1) and had the lowest 18-week body weight and breast angle (Table 3). Phenotypic and genetic correlations between the body traits and incidence of DMS were estimated in the same populations from which
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
GENERATION
FIG. 1. Response to selection for incidence of degenerative myopathy of the Musculus supracoracoideus (DMS) plotted as deviations from incidence recorded for the control strains after percentages had been transformed by arcsin yjpercentage. The H, C, and L represent strains selected for high, with no selection, and selected for low incidence of DMS, respectively, in each of Bases E and F.
423
DEGENERATIVE MYOPATHY IN CHICKENS
DISCUSSION
There were two different methods of selection for incidence of DMS, namely, pedigree selection on DMS incidence after exercise for producing Generations 1 and 2, and mass selection for the related trait plasma CK for Generation 3. No direct selection for DMS could be practiced because necropsy was necessary for diagnosis. Response to selection for DMS was low or negligible, probably due to the erratic and low selection differentials caused by the indirect selection. When the selection was expressed in terms of DMS incidence of the parents themselves, which was determined only retrospectively after the parents were reproduced, killed, and necropsied, the parental S sometimes differed dramatically from those for the breeders' parents, not only in magnitude
TABLE 3. Least squares means and significance of differences in body weight and breast angle after 3 generations of selection for degenerative myopathy of the Musculus supracoracoideus (DMS)
Source of variation
No.'
18-wk Body weight
8-wk Body weight N"g>
~
18-wk Breast angle
o
*
***
558 1131
220.5 217.1
334.4 312.4
Sex Male Female
***
** *
***
749 940
243.6 193.9
361.1 285.6
92.5 97.8
Base E F
***
***
897 792
219.8 217.7
329.5 317.3
94.3 96.1
Strain High Control Low
*
*
* **
418 725 546
217.5 220.6 218.2
324.1 326.7 319.3
96.3 95.6 93.6
Necropsy class Affected Nonaffected
*** 97.4 92.9
1 Only birds with a recorded DMS status when killed at 21 weeks of age and measured for all three production traits are included.
*P<.05. ***P<.001.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
but also in sign. For example in Strain FL, in Generation 0, breeders' S values were +2 and +6 for sires and dams, respectively, but were —51/ —46 for sires' and dams' parents (Table 1). The greatest selection response (Fig. 1) was observed in Generation 3, indicating that indirect selection on the basis of plasma CK response to exercise could be effective. Selection in the downward direction in the L strains was most successful in the EL strain, the E strains having a higher DMS incidence in general, and thus a greater potential for improvement, than the F strain. Results of selection in Generation 3 indicated that selection for a continuous trait, plasma CK, related to DMS incidence and measured in chickens to be selected, may be more effective than selection for the threshold trait of interest, DMS, measured only in parents of the chickens to be selected. In direct selection on CK levels, the S in terms of DMS incidence may not be an adequate description of the selection practiced, because the level of CK may have been an indication of DMS susceptibility even though the DMS itself may not have been expressed in
heritability estimates were reported above. Phenotypic correlations, shown in Table 2, were all positive. The comparable genetic correlations (not shown) were much more variable in that they ranged from —1.26 to .54 with about half of them being positive.
424
HOLLANDS ET AL. ACKNOWLEDGMENTS
The authors wish to thank P. Griffin for skillful technical help and C. Williams for discussion of statistical procedures. REFERENCES Becker, W. A., 1975. Manual of Quantitative Genetics. Univ. Press, Washington State Univ., Pullman, WA. Dempster, E. R., and I. M. Lemer, 1950. Heritability of threshold characters. Genetics 35:212—236. Dickerson, G. E., 1960. Techniques for research in quantitative animal genetics. Pages 57—96 in Techniques and Procedures in Animal Production Research. Am. Soc. of Anim. Prod., Champaign, IL. Dickinson, E. M., J. O. Stevens, and D. H. Heifer, 1968. A degenerative myopathy in turkeys. Page 6 in Proc. 17th West. Poult. Dis. Conf., Univ. Calif., Davis, CA. Falconer, D. S., 1976. Pages 3 0 1 - 3 1 1 in Introduction to Quantitative Genetics. The Ronald Press Company, New York, NY. Grunder, A. A., K. G. Hollands, J. S. Gavora, J. R. Chambers, and N.A.G. Cave, 1984. Degenerative myopathy of the Musculus supracoracoideus and production traits in strains of meat-type chickens. Poultry Sci. 63:781-785. Harper, J. A., P. E. Bernier, D. H. Heifer, and J. A. Schmitz, 1975. Degenerative myopathy of the deep pectoral muscle in the turkey. J. Hered. 66: 362-366. Harper, J. A., P. E. Bernier, and L. L. ThompsonCowley, 1983. Early expression of hereditary deep pectoral myopathy in turkeys due to forced wing exercise. Poultry Sci. 62:2303-2308. Harper, J. A., D. H. Heifer, and E. M. Dickinson, 1971. Hereditary myopathy in turkeys. Page 76 in Proc. 20th West. Poult. Dis. Conf., Univ. of Calif., Davis, CA. Hollands, K. G., A. A. Grunder, and J. S. Gavora, 1983. Plasma creatine kinase as an indicator of degenerative myopathy of the M. supracoracoideus (DMS) in exercised chickens. Poultry Sci. 62:1435. (Abstr.) Hollands, K. G., A. A. Grunder, C. J. Williams, J. S. Gavora, J. R. Chambers, and N.A.G. Cave, 1981. Degenerative myopathy in meat-type poultry: its effect on production traits in chickens and its identification in live turkeys. Pages 337—344 in Quality of Meat. Proc. 5th Eur. Symp., Apeldoorn, The Netherlands. Richardson, J. A., J. Burgener, R. W. Winterfield, and A. S. Dhillon, 1980. Deep pectoral myopathy in seven-week-old broiler chickens. Avian Dis. 24: 1054-1059. Robertson, A., and I. M. Lemer, 1949. The heritability of all-or-none traits: viability of poultry. Genetics 34:395-411. Siller, W. G„ L. Martindale, and P.A.L. Wight, 1979a. The prevention of experimental deep pectoral myopathy of the fowl by fasciotomy. Avian Pathol. 8:301-307. Siller, W. G., P.A.L. Wight, and L. Martindale, 1979b.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015
the bird. The CK level, a continuous variable, could be regarded as an expression of the underlying continuous character, similar to that in the general concept of threshold characters (Falconer, 1976). Incidence of DMS can then be viewed as the all or none trait expressed only when the continuous underlying character exceeds a threshold. The heritability estimates for incidence of DMS were quite variable (Table 2). The Chisquare method of estimating heritability from binomial data was based on half-sib relationships, and thus conversion to the normal scale is reasonable, according to Van Vleck (1972), when the proportion of affected individuals is not too small and normal heritability is not too large. The heritability estimate of .48 is at least medium, and based on this heritability, and assuming birds affected by DMS can be identified accurately when alive, a breeder can expect progress in selecting against DMS. Induced wing exercise and measurement of plasma CK (Hollands et al, 1983) should enable him to select breeders more accurately than if he uses family selection. In any selection program for improving meat-type chickens, growth rate and conformation as well as other traits are most important. Data of Tables 2 and 3 suggest that the desirable traits such as fast growth rate and large breast angle are greater in DMS positive than DMS negative birds. These observations are consistent with those of Hollands et al. (1981) and Grunder et al. (1984) who presented evidence that faster-growing birds with good conformation were more susceptible to DMS than slowergrowing birds with bad conformation. Also in agreement were the results of Siller et al. (1979b) who were able to cause DMS in turkeys and broiler chicken stocks but not in slower growing, narrow breasted, egg-type chickens by forced wing exercise. The phenotypic correlations of Table 2 also support these observations. Selection for body weight and conformation alone would lead to DMS susceptibility (predisposition) as apparently happened in commercial broiler chicken stocks (Grunder et al, 1984). Selection against DMS alone may lead to reduction in body weight and conformation. Therefore, a multitrait approach, combining selection for rapid growth and body conformation with selection against DMS susceptibility, would be best for improving meat type stocks.
DEGENERATIVE MYOPATHY IN CHICKENS
Exercise-induced deep pectoral myopathy in broiler fowls and turkeys. Vet. Sci. Commun. 2: 331-336. Steel, R.G.D., and J. H. Torrie, 1960. Principles and
425
Procedures of Statistics. McGraw-Hill, New York, NY. Van Vleck, L. D., 1972. Estimation of heritability of threshold characters. J. Dairy Sci. 55:218-225.
Downloaded from http://ps.oxfordjournals.org/ by guest on April 9, 2015