- Email: [email protected]
Lymphoid Leukosis Virus Infection: Effects on Production and Mortality and Consequences in Selection for High Egg Production1
BREEDING AND GENETICS Lymphoid Leukosis Virus Infection: Effects on Production and Mortality and Consequences in Selection for High Egg Production1 J. S. GAVORA, J. L. SPENCER,2 R. S. GOWE, and D. L. HARRIS 3 Animal Research Institute, Agriculture Canada, Ottawa, Ontario, Canada K1A OC6 (Received for publication October 29, 1979)
1980 Poultry Science 59:2165-2178 INTRODUCTION L y m p h o i d leukosis (LL) is a virus-induced lymphoblastic malignancy originating in t h e bursa of Fabricius. Chickens are infected either congenitally, t h r o u g h t h e egg (vertical transmission), or by direct c o n t a c t with infected chickens (horizontal transmission). Pathogenesis of L L has been described by Siccardi and Burmester ( 1 9 7 0 ) a n d b y C o u d e r t et al. ( 1 9 7 7 ) . Following infection of susceptible chickens, large a m o u n t s of virus are p r o d u c e d in m a n y tissues w i t h o u t any a p p a r e n t harmful effects.
1
Animal Research Institute Contribution No. 890. Animal Pathology Directorate, Health of Animals Branch, Agriculture Canada, Animal Diseases Research Institute, Nepean, Ontario, Canada K2H 8P9. 3 Animal Genetics Research Unit, USDA, SEA, AR, and Department of Animal Sciences, Purdue University, W. Lafayette, IN 47907. 2
Follicles of t h e bursa of Fabricius m a y b e c o m e t r a n s f o r m e d a n d remain quiescent until sexual m a t u r i t y of t h e chicken (16 t o 22 weeks) w h e n t h e cells start divisions, acquire neoplastic characteristics, and spread t o o t h e r visceral organs where t u m o r s m a y develop. Mortality usually does n o t occur until chickens have reached sexual m a t u r i t y . T w o levels of genetic resistance t o LL are recognized (Crittenden, 1 9 7 5 ) : 1) cellular resistance t o virus infection which is simply inherited b u t is very specific for t h e subgroup of virus, and 2) resistance t o t u m o r d e v e l o p m e n t in L L virus ( L L V ) infected birds which is m o r e c o m p l e x a n d appears t o be less subgroup specific. Mortality from L L is generally low. Purchase et al. ( 1 9 7 2 ) surveyed m o r e t h a n 1 4 5 , 0 0 0 birds over a period of a p p r o x i m a t e l y 18 m o n t h s . T h e y r e p o r t e d average L L m o r t a l i t y of 3.3 and 4.6% in Marek's disease vaccinated and nonvaccinated birds, respectively. Lifetime losses
2165
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
ABSTRACT Lymphoid leukosis (LL) is a virus-induced, lymphoblastic malignancy of chickens that can be congenitally transmitted. Mortality from LL is generally low. Effects of LL virus (LLV) on production and mortality were investigated in approximately 2000 Leghorn pullets in each of two consecutive years. The pullets were from nine strains developed in Ottawa, of which three were unselected control strains and six were strains under selection for up to 27 generations for high egg production and a complex of related commercially important traits. The overall frequency of birds shedding LL virus or gs antigen into eggs (LL-S) was significantly lower in the selected strains (3.9%) than in the control strains (18.5%), indicating that LLV may have negative effects on production and cause elimination of LL-S birds by selection. Such significant effects were indeed detected: the LL-S pullets produced to 497 days of age in 1976 and 1977, respectively, 30 and 25 eggs less per hen-housed than the nonshedders. The LL-S birds matured sexually later, produced smaller eggs at a lower rate, and their eggs had a lower specific gravity, indicating thinner shells. Mortality from all causes to 497 days was significantly higher in LL-S birds (+14.8%) in 1976. In 1977 the increase (+5.5%) did not reach statistical significance. In both years the mortality from LL itself remained very low. In another study, eggs from one of the control strains were incubated and hatched when the dams were 291 and 483 days old. The eggs from LL-S dams had 2.4% lower fertility and 12.4% lower hatchability. The effects on hatchability were more pronounced in the older dams. Since the lower production of LL-S birds results in a lower frequency of such birds in strains selected for high egg production, it is suggested that a part of the difference between the performance of the selected and control strains (Ag) is due to reduction in the frequency of LL-S birds ( A L ) rather than due to true genetic gain. In this study, the size of A L relative to Ag was estimated at 4 to 14% for egg production and 3 to 7% for egg weight. The negative effects of LLV infection on egg production, mortality, hatchability, and genetic gains show the desirability of producing chickens free of LLV infection. (Key words: lymphoid leukosis, egg production, mortality, selection, genetic gain)
2166
GAVORA ET AL.
MATERIALS AND METHODS
Strains of Chickens and Management of the Flocks. The strains tested for LLV infection have been previously described (Gowe, 1977; Gowe et al, 1973; Gavora et al, 1975) and the basic data on the strains are summarized in Table 1. Management of the flocks tested for LLV infection has been described by Gowe (1977), Gowe et al. (1973), and Spencer et al. (1979). Briefly, the chickens were reared in three-tiered group cages with sheet metal partitions between cages. Cages were randomly assigned to strains. Within each strain, day-old pullets from each sire family were randomly distributed through the cages. Approximately 75 chicks were placed in the top-tier cages on the day of hatch and were later divided among the three tiers. At 20 weeks, the pullets were distributed in a stratified random manner into individual 20 cm stairstep cages in two laying houses. The birds were fed mash rations ad libitum throughout the test. In both 1976 and 1977, all birds in the laying houses were divided into two groups that were fed slightly different diets: in 1976, one half of the birds (nutrition treatment 1 in Tables 3 to 5) received a diet with .4% phosphorus, while the ration fed to the other half (nutrition treatment 2 in Tables 3 to 5) had .35% of phosphorus. Similarly, in 1977, birds on nutrition treatment 1 (Tables 3 to 5) were fed a layer ration with 3.1% calcium, while birds on nutrition treatment 2 were given
free choice of oyster shell, spread on top of the same ration fed the other half of the birds. Individual egg production was recorded 5 days a week and then converted to a 7-day week basis. The birds were weighed at housing, at 1 year of age, and at the end of the egg production test (497 days). Egg weight and egg quality traits were measured on eggs laid within 5 consecutive days when the birds were approximately 240 and 450 days old. Collection of Egg Samples and Tests for LLV Infection. In 1976, eggs were collected from all hens of strains 5, 1, 4, 7, and 8 that were in one of the two laying houses between 255 and 259 days and between 283 and 287 days of age (Spencer et al, 1979). In 1977, eggs were similarly collected between 242 and 250 days of age from pullets of all strains in one of the two laying houses, but the size of the populations sampled was equalized to 220 pullets per strain. Results from LLV tests on one egg from each of 1935 and 1848 hens were used in the evaluation of LL effects on production in 1976 and 1977, respectively. To assess the consistency with which hens shed LLV into eggs, a second egg from each of 1394 hens was collected in 1977 when the hens were 468 to 476 days old. Detection of LLV and the group specific (gs) antigen in the albumen of unincubated eggs was reported by Spencer et al. (1976, 1977). In this study, the phenotypic mixing test (Okazaki et al, 1975) was used to detect LLV in the albumen of the eggs collected in 1976. The microcomplement fixation test (Sever, 1962; Sarma et al, 1964) was used to detect the gs antigen in the albumen of eggs collected in 1977. The gs antigen is an internal component of the virus common to all avian leukosissarcoma viruses. There is generally good agreement between results of tests for virus and gs viral antigen (Spencer et al, 1979). For the purposes of this presentation LLV shedders (LL-S) are birds that produced eggs in which LLV or gs antigen were detected. Examination of Effects of LLV Infection of the Dam on Fertility and Hatch ability. The hens of strain 5 tested for LLV infection in 1977 were pedigree mated by artificial insemination (1 male to 8 females) to a random sample of strain 5 males. Eggs were collected for incubation over two periods of two weeks each and chicks were hatched when the parents were 291 days old (hatch 1) and 483 days old (hatch 2). The eggs collected in the first week
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
from LL amounted to 1.2 to 3.4% in experimental strains of Leghorns and meat-type breeders maintained on the same premises as the birds in the current study (Gavora et al, 1975). In large field surveys conducted in Great Britain, 1.4% of total laying house mortality was attributed to LL in laying flocks (Randall et al, 1977) and 1.5% in broiler breeder flocks (Jones et al, 1978). Occasionally much higher losses from LL are experienced in some flocks (Crittenden and Witter, 1978). Evidence for negative effects of subclinical infections with LLV has been reported (Spencer et al, 1978; 1979; Chase, 1978). The main objective of the present study was to assess the economic losses from LL in terms of both mortality and effects of subclinical LLV infection on performance in egg production stocks. Consequences of such effects in strains selected for high egg production were also evaluated.
1959
Four commercial stocks crossed
Derived from strain 7
Derived from strain 7
Four commercial stocks crossed
C
s
s
c
7
9
8
10
1973
1969
1969
1969
c = unselected control strain; S = strain selected
a
1951
Seven Canadian R.O.P. unrelated stocks crossed
S
4
Derived from strain 4
S
2
1950
1971
Same as strain 5
S
3
1950
Derived from Strain 3
S
1
Narrow genetic base (Ottawa)
O
Year
Base population
Origin
5
Name Type
Sitrain
for high egg production.
None
None
478
1123
High hen-housed egg production
1130
475
Fertility, hatchability, viability, egg size, egg quality
None
High egg production Fertility, hatchability, rate viability, egg size, egg quality
None
1119
1128
High hen-housed egg production Fertility, hatchability, viability, egg size, egg quality
High egg production Fertility, hatchability, viability, egg size, rate egg quality
High hen-housed egg production 1123
476
1127
_
Fertility, hatchability, viability, egg size, egg quality
None
Secondary traits
Housed in laying house
High egg production Fertility, hatchability, rate viability, egg size, egg quality
None
Primary trait
Selection criteria
No. o
TABLE 1. Description of the strains, number of pullets housed, and incidence of shedding lymphoi
rom http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
GAVORA ET AL.
2168
u s
a
d
i
j:i
k:j:i "1
ah
e
lj:i
ijklm
= population mean = random effect of sire = fixed effect of LL status of dam within sire = random effect of dam within LL status within sire = fixed effect of hatch = interaction of LL status of a dam within sire with hatch = random individual effects
All-or-none data, such as mortality, fertility, or hatchability of an egg and classification of a newly hatched chick as normal or crippled were coded " 1 " for birds or embryos that died, for infertile eggs or eggs that did not hatch, and for chicks classified as crippled and " O " for the remaining eggs or birds. Harvey's (1972) LSMLMM program was used for the least squares analyses of the data. Linear model approach to the analysis of categorical data has been presented by Schaeffer and Wilton (1976) and by Grizzle et al. (1969) and was previously used for analysis of all-or-none data by Gavora etal. (1977).
Yijkim = u + a j + s j ; i + n k + b, + an i k RESULTS AND DISCUSSION
+ abil+nbkl+eijklm, where u a
S
i
j:i
n
k
b, an i k , ab n , and nbkl e
ijklm
= = = = =
population mean fixed effect of strain random effect of sire within strain fixed effect of nutrition treatment fixed effect of LL status of the individual
= interactions = random individual effects.
The statistical analyses of the strain 5 fertility anid hatchability data were performed
Yijklm =
u + s
i
+ a
j:i
+ d
k:j:i
+ h
l
+ ahjj.j + e i j k l m ,
where yijklm
=
f e r t m t y » hatchability of an individual egg or the classification of a chick as normal or crippled
Consistency of Shedding LLV into Eggs. Tests for shedding gs antigen into egg albumen were conducted on each of two eggs gathered several months apart from hens hatched in 1977. Only in 15 hens (1.1%) out of the 1,394 hens tested twice did the results of the second test differ from the results of the first test. In 5 instances, birds previously tested positive tested negative for gs antigen in the second test, while in 10 instances birds negative on the first test became positive on the second test. These results indicate 1) that gs antigen was shed into egg albumen with a high degree of consistency in the populations tested and 2) that the spread of LLV among adult hens (if any) did not increase the proportion of shedders significantly. Frequency of LL-S Birds. Strains under selection for high egg production had a significantly lower incidence of LL-S birds than unselected control strains (Table 2). Within the selected strains, breeders had been selected on the basis of their egg production and related traits but without paying any attention to their LLV status. In the control strains, breeders had been selected by a random process, completely disregarding both their egg production and LLV status. The lower frequency of LL-S in the
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
of each egg save period were stored in cryovac bags flushed with nitrogen to reduce the effects of egg storage on hatchability. All eggs were candled on day 18 of incubation and eggs with dead embryos and infertile eggs were identified and removed. On the day of hatch, all chicks hatched were examined and those with incompletely closed navel or with external malformations, as well as extremely weak chicks, were classified as "crippled". Statistical Analyses. The frequency of LL-S birds (Table 1) in strains 2 and 9 in 1977 was very low (1 and 2 LL-S birds in strains 2 and 9, respectively) and, therefore, these strains were excluded from the statistical analyses. The average number of pullets tested for shedding LLV into eggs per family of full sisters was 2.2 in 1976 and 1.4 in 1977. Thus, the infection status of the birds was often confounded with dam family, and therefore, the effect of dams was not included in the statistical analyses of the production and mortality traits. The following statistical model was used in the analyses:
LYMPHOID LEUKOSIS AND EGG PRODUCTION
2169
TABLE 2. Comparison of the frequency of birds shedding lymphoid leukosis virus (LLV) into eggs in selected and control strains and among breeders and the remaining birds that were not used in the reproduction of the strains 1976 Type of birds
Type of strains
Breeders
C*
Number tested 218
1977
Percent LLV Chishedders square 23.2
361 17.2 2.53
S
176 348
16.5 1.7
C
1193 394
5.4 20.0
S
1541
4.1
579
19.3
405 619
17.3 2.3
2180 984
4.7 18.5
2799
3.9
.67
.05 229 17.9 271 3.0
8.50'* Others All birds
Num- Percent ber LLV Chitested shedders square
.49 987 3.9 590 17.5
117.60**
7.30**
103.20** 1258
3.7
218.63**
C = Unselected control strains; S = strains selected for high egg production. For more details see Table 1. **P<.01.
selected than in the control strains (Table 2) suggests that production is negatively influenced by LLV infection. The lower production in LL-S birds then results in their elimination in the process of selection for high egg production. This is supported by the comparison of the frequency of LLV shedding among breeders and other birds not used in strain reproduction (Table 2). In the selected strains, the frequency of LL-S birds was significantly lower among breeders than among other birds in 1976 and also in the 1976 and 1977 data combined. However, there was no significant difference in the frequency of LLV positive birds among breeders and other birds within the control strains. These findings are consistent with the earlier conclusion (Spencer et al., 1979), based on the 1976 data, that strains under selection for high egg production have generally a lower incidence of LL-S birds than corresponding control strains. Of particular interest was control strain 10 (Table 1) synthesized in 1973 and maintained since unselected. The percentage of LL-S birds in this strain was in general agreement with results from the other two control strains indicating that, under the conditions of this study, the percentage of LL-S birds in populations maintained without selection tends to settle around 20%. It should be noted that the reduction in the percentage of LL-S birds in the selected strains cannot be attributed to an increase in genetic
cellular resistance to LLV infection. As shown by Spencer et al. (1979), the selected strains were at least as susceptible to LLV infection as were the control strains. The data on the percentage of LL-S birds are also in agreement with previous reports (Spencer et al., 1978, 1979; Chase, 1978) that LL-S birds tend to be eliminated in selection for high egg production. Comparison of Production and Mortality of LLV shedders and nonshedders. Tables 3 to 5 contain the estimated mean performance of LLV nonshedder birds in the selected and control strains tested, the effects of LLV shedding in terms of mean differences between LLV shedder and nonshedder birds, and degrees of freedom and mean squares from the analyses of variance. Large and significant effects of LLV shedding on hen-housed egg production were detected in both 1976 and 1977 (Table 3). It should be noted that, as mentioned above, the periods over which eggs were collected for tests of LLV shedding were from 255 to 287 days in 1976 and from 242 to 250 days in 1977. Thus, all mortality in this study was only from the time of the tests and consequently the hen-housed egg production from 141 to 273 days had little mortality component. The reduction of hen-housed egg production was largest towards the end of the 1-year egg production period (Table 3) and its size was quite consistent in the 2 years. No significant strain by LL status interaction
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
Others Breeders
Combined
Num- Percent ber LLV Chitested shedders square
C
7 8
6 334 1 1 6 6 1 1069
4 231 1 1 4 4 1 1688
Strain (St) Sire/strain Nutrition treatment (N) LLV Status (L) St X N StX L NX L Residual
13,100** 280** 229 2,164** 35 209 555 145
104/-3 105/-8**
94/-9** 110/-9** 911-1
9,321** 274** 4 1,433** 87 133 87 190
105/-6** 104/-4
3,117* 1,050** 83 15,113** 515 1,675 0 617
68/-15** 68/-15**
259A-27** 259/-33**
2,564** 832 1,541 13,075** 199 436 4,138* 717
**P«.01.
*P«.05.
Tests for LLV shedding were performed on eggs collected between 255 and 287 days in 1976 and between 292 in
37,895** 3,100** 1,005 58,938** 2,114 6,369 957 1,895
Mean squares
65/-22** 62/-7
66/-10
231/-18 274/-49**
58/-9 71/-25*
58/-4 73/-24**
93/-6 115/-10
270/-54*
60/-20*
114/-7
115/-7* 67/-26**
66/-21** 76/-1
88/-5 114/-2 111/ 3
88/-1 114/-4
236/-33** 281/ 4
259/-30**
64/-14** 57/-13 77/-20* 59/-5
68/-15**
105/-5**
104/-6**
C = Unselected control strain, S = strain selected for high egg production. For more details see Table 1.
1977
1976
Degrees of freedom
1 2
10
S
4
s c
C S S
5 1 3
a
of variation
Nutrition treatment
Strain
All birds
141-49
1976
1977
386-497 days 1976
1977
1976
141-273 days
Hen-housed egg production
Mean of LLV nonshedders/Difference LLV shedders
TABLE 3. Least squares mean hen-housed egg production and mortality for LLV nonshedders and LL effect expres nonshedders; degrees of freedom and mean squares for all birds tested for shedding lymphoid leuko
d from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
C S
c
7 8
10
4 228 1 1 4 4 1 1564
Strain (St) Sire/strain Nutrition treatment (N) LLV status (L) StX N StX L NX L Residual
6 330 1 1 6 6 1 1015
1977
160.5/ 6.9* 147.7/ 5.6
163.2/ 1.9 150.9/ 5.9**
7,203** 201** 97 436** 35 144 251 74
157.1/ 1.2 156.4/ 4.7**
149.1/
150.2/-1.8
5,073** 163** 286 601** 56 181* 96 86
154.8/ 2.0 155.6/ 4.4*
161.1/ 8.9**
1.2
168.1/ 4.0 151.4/-1.7 148.5/-2.3
155.2/ 3.2**
168.9/ 4.8* 150.4/ 4.2
156.7/ 3.0*
1,194** 184** 41 863** 94 528** 33 107
**P«.01.
*P«.05.
75. 73.
73.
69. 78.
76.
67.4 83. 73.
74.
197
1 3 1
1,9
1,1 2
Mean squares
76.8/ - 4 . 8 * 76.9/ - 3 . 5
69.5/ 1.4 78.7/-11.7**
78.3/ - 7 . 7 * *
74.3/ - 7 . 6 * * 83.6/ 4.8
76.9/ - 4 . 2 * *
1976
Hen-day egg product from first egg to 497 day:s (%)
C = Unselected control strain; S = strain selected for high egg production. For more details see Table 1.
1976
Degrees of freedom
S
4
1 2
Ca S S
5 1 3
Source of variation
Nutrition treatment
Strain
All birds
1977
1976
Age at the first egg (days)
Mean of LLV nonshedders/Difference LLV sh
TABLE 4. Least squares mean age at sexual maturity, hen-day egg production, and survivors egg production for difference between shedders and nonsbedders; degrees of freedom and mean squares for birds that (497 days) and were tested for shedding lymphoid leukosis virus (LLV) into
from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
4 6 228 330 (224)b (324)b 1 1 1 1 6 4 6 4 1564 1015 b (1409) b (825)
Strain (St) Sire/strain
25 90** 5 12 5 10
567** 25** 71* 101** 3 23 34** 11
337** 19**
55.9/ - . 6 55.6/-2.0**
.9
120** 50 13 16 87** 17
329** 46**
63.8/ .1 63.6/-2.3*
-.5
2.2
19
90*
178** 134** 22 31
141** 34**
Degrees of freedom in the analyses of the 450-day data.
**P<.01.
*P«.05.
0 122** 14 8 3 14
186** 37**
Mean squares
63.5/ - . 4 62.9/-3.1**
862/859/-
869/869/-
64.0/-2.0
829/-
61.0/-1.0 62.8/-1.9
862/ 874/
860/-
1976
64.3/-3.4
65.8A-2.9
65.4/-3.1*
59.0/
63.2/-1.7**
C = Unselected control strain; S = strain selected for high egg production. For more details see Table 1.
1977
1976
of variation
Degrees of freedom
56.4/-1.1* 56.1/ 1.6**
56.6/-1.6
62.2/ 65.4/
10
64.8/
53.7/ - . 6 56.2/ .8
54.6/-1.3 58.0/ - . 9
.2
57.6/
7 8
Nutrition treatment
Nutrition treatment (N) LLV status (L) StX N St X L N XL Residual
63.7/-1.1
1977
at 450 days
61.5/-2.1 64.7/-2.9
4
s s
55.8/-1.3**
56.2/-1.3** 51.5/ - . 4 57.2/ - . 1 57.7/-2.6* 51.4/-4.6**
Ca
1976
53.5/-2.5** 57.4/-2.1
5 1 3
1 2
Strains
All birds
1977
Egg weight (g)
1976
at 240 days
Mean of LLV nonshedders/Difference LLV shedd
TABLE 5. Least squares mean egg weight and egg specific gravity at 240 days and 450 days of age for LLV nonshe between shedders and nonshedders; degrees of freedom and mean squares for birds that survived to th were tested for lymphoid leukosis virus (LLV) shedding into eggs
from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
LYMPHOID LEUKOSIS AND EGG PRODUCTION
were observed in eggs from both younger and older birds. Body weights (not shown in the tables) of the LL-S birds tended to be slightly lower; differences, ranging between 30 and 60 g per bird, were not significant. Also, differences in Haugh units and in the percent of eggs with blood spots (not shown in the tables) were generally small and were mostly not significant. Differential responses of strains to LLV infection were indicated by significant strain by LLV status interactions in hen day percent and survivors' egg production in 1976 and in sexual maturity and 240 day egg weight in 1977. Significant nutrition treatment effects were found for egg weight and egg specific gravity. Significant nutrition treatment X LLV status interactions were found only in 1977 hen-housed egg production. The nutrition treatments will be dealt with further in a separate report, where the performance of all the birds, not just those that were tested for LLV shedding, will be evaluated. Table 6 shows the major causes of laying house mortality. An increase in mortality from all causes among the LL-S birds is indicated by the data. No deaths were attributed to LL in 1976; in 1977, mortality from LL was very low. The . 1 % mortality attributed to LL in the 1977 LL-S birds can be explained by the fact that in this study LL-S birds were identified by tests on egg albumen and, thus, some birds infected with LLV but which did not shed the virus into eggs may have been classified as LLV nonshedders. The total mortality in Table 6 is slightly different from that in Table 3. This is because Table 6 contains raw percentages while
TABLE 6. Association of increased laying bouse mortality percent with shedding lymphoid leukosis virus (LLV) into eggs Lymphoid leukosis status a
Cause of death
1976 LLV nonLLV shedders shedders
LLV nonshedders
LLV shedders
Reproductive disorders Marek's disease Fatty liver syndrome Lymphoid leukosis Other causes Total mortality
1.7 1.2 .7 0 1.4 5.0
1.8 1.9 .6 .1 1.5 5.9
3.4 3.4 1.3 1.3 3.5 12.9
4.0 3.3 2.6 0 5.5 15.4
1977
Based on the detection of lymphoid leukosis virus or the gs antigen in eggs collected between 255 and 287 days in 1976 and between 242 and 250 days in 1977.
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
is indicated by the analyses. However, the data in Table 3 suggest that strain 3 may have a better ability to tolerate LLV infection than the other strains. In 1976, laying house mortality was significantly higher (+14.8%) in the LL-S birds (Table 3). In 1977, mortality among the LL-S was also higher, but the difference (+5.5%) did not reach significance (P>.05). The absence of significant strain by LLV status interaction suggests that with the possible exception of strain 3, the effect of LLV shedding on the bird's survival was similar in all genotypes (strains) tested. As mentioned earlier, only mortality from the time of the LLV shedding tests was considered in this study. Since eggs for such tests were collected no earlier than at 242 days, the data in Table 3 do not include either early laying house mortality or rearing mortality. Hence, the effects of LL on mortality may be underestimated. As shown in Table 4, LLV shedding was associated with delayed sexual maturity. The mean effect was a delay of approximately 3 days, but the size of the effect varied widely among the individual strains. Egg production rate and the number of eggs produced by the birds that survived to 497 days was also lower for the LL-S birds, indicating that infections that caused LL-V shedding but did not result in the death of the birds were associated with reduced egg production. LL-S birds produced smaller eggs (Table 5) and their eggs had a lower average specific gravity, indicating thinner egg shells which could lead to higher egg breakage. Such effects
2173
2174
GAVORA ET AL.
physiological processes. These effects ranged from susceptibility to infectious diseases (Marek's disease) through deaths from reproductive disorders and fatty liver syndrome to effects on fertility, hatchability, sexual maturity, egg production rate, egg weight, and egg shell strength. A majority of the effects was associated with subclinical infection with LLV because mortality from LL was very low (Table 6). Although the association of LL-S status with lower production and higher mortality was clearly demonstrated in this study, the possibility that the association was not due to a direct effect of LLV cannot be eliminated. If, for example, shedding of LLV into eggs occurred primarily among hens less competent immunologically, then their low production and high mortality may have been the result of the lower immunological competence rather than the effect of LLV. Mechanisms through which the subclinically infected birds may be affected by LLV have not been investigated in this study. However, it is noteworthy that in electron micrographic studies DiStefano and Dougherty (1968, 1969) found LLV in the ovary and testis, pituitary, thyroid, parathyroid glands, and in the adrenal cortex of LLV infected chickens that appeared healthy. This suggests possible effects of LLV on a wide range of endocrine functions and may at least partly explain the negative effects on economic traits observed in this study. Research on the immunosupressive effects of LL was recently reviewed by Sharma (1979). He concluded that most researchers noted some depression in antibody synthesis by LLVinfected chickens. Although the exact mechanism of this impairment is unknown, it may be related to the affinity of LLV to the bursa of Fabricius. In vitro studies indicated that LLV infection also lowers cellular immune competence. It appears that immune response may be influenced by LLV infection even without the accompanying cellular transformation. Thus, immunosupression associated with LLV infection may explain the increased mortality from causes other than LL observed among the LL-S pullets in this study. Effects of LL on Selection Gains. As discussed earlier, several aspects of production were negatively influenced by LLV shedding. It has also been demonstrated that the frequency of LL-S birds is reduced in strains under selection for high egg production. From these
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
least squares means of total mortality are shown in Table 3. Effects of LLV Infection on Fertility and Hatchability. Mean fertility, early and late embryonic mortality, hatchability, and percentage of crippled chicks for LLV nonshedder dams, as well as mean differences in these parameters between the LLV shedder and nonshedder dams, are shown in Table 7, together with degrees of freedom and mean squares. Significant decrease in fertility, hatchability from fertile eggs, and consequently in hatchability from eggs set was associated with LLV shedding by the dam. The decrease in hatchability of fertile eggs was primarily due to higher mortality of embryos prior to day 19 of incubation. Since the eggs classified as infertile by candling were not broken and examined for presence of zygotes, it is possible that some of the reduction in fertility might have been due to early embryonic mortality not distinguishable by candling. The increase, by 1.3%, in the percentage of crippled chicks among the progeny of LL-S dams, although substantial relative to the 1.9% crippled chicks observed in the progeny of negative dams, did not reach statistical significance. Both hatchability and percentage of crippled chicks were affected by hatch. Since the insemination, egg save, and incubation procedures were similar in both hatches, hatch effects were likely primarily associated with age of the brids. Thus the observation of more severe effects of LLV infection on hatchability and percentage of crippled chicks in Hatch 2 (when the dams were 487 days old; Table 7) coincides with the observation that LLV infection had the largest effect on egg production in the last part of the production period (386 to 497 days; Table 3). Although in the study on hatchability only one strain of chickens was used, the size of the effect of LLV status of the dam on production of day-old chicks (hatchability of eggs set) is large enough to be of concern and to warrant more investigation. In selected strains, the reduced hatchability could also be expected to reduce the contribution of progeny from LL-S dams that have not been eliminated by selection. The above data on the association of LLV shedding with lower production and higher mortality from all causes suggest the effects of LLV on a number of distinctly different
Degrees of freedom
1 2
2,000
3,063** 1,271** 10,871** 2,148** 529
20
20 105 1 14 1732
1,683** 780 150 525 511
Mean squares
6.2/ 5.4 8.9/ 17.1
20 103 1 14 1611
20
Degrees of freedom
7.6/ 11.2**
33,533**
Mean squares
Degrees of freedom
89.3/- -2.4**
90.2/- -1.1 88.5/- -3.9
**P«.01.
1,692** 757** 3,134** 1,815** 347
1,769
Mean squares
3.1/ 9.0 14.6/-3.6
10.2/ 2.8**
19-21 days
Embryonic mortality (%)
1-18 days
Significance of the effect of the dam's LL status was not tested within hatch.
Sire (S) 20 LLV status of dam (Ds)/S 21 Dam (D)/Ds/S 109 Hatch (H) 1 H X Ds/S 14 Residual 1947
Source of variation
Hatch*
All dams
Fertility (%)
20 105 1 14 1732
20
Degrees of freedom i
6,322** 2,654** 41,862** 57,905** 990
5,439
Mean squares
83.9/ - 5 . 6 78.5/-19.2
81.2/-12.4**
fertile eggs
Hatcha
Mean of LLV nonshedder dams/Difference LLV shedder dams
TABLE 7. Least squares mean fertility, embryonic mortality, hatchability, and percentage of crippled chicks for as difference between shedder and nonshedder dams; degrees of freedom and mean squares from
from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
2176
GAVORA ET AL.
where Ag is the difference in performance between a selected and a control strain, AQ'IS the genetic gain, and AL is the gain due to reduction in the frequency of LL-S birds resulting from selection. These two parts can be
estimated as: AG = Xi -x2, and A
L
=
q2(x2-y2)-qi(xi-yi),
where ~k~t and ~x2 are mean performances of the birds that did not shed LLV in the selected and control strains, respectively, y"t and y 2 are the respective mean performances, and q ! and q 2 are the respective frequencies of LL-S birds in the selected and control strains. Table 8 shows examples of Ag, AQ, and AL calculated for egg production and egg weight for control strain 5 and corresponding selected strains 1 and 3, and for control strain 7 and corresponding selected strains 9 and 8. The calculations were based on 1977 data. The size of the improvement in production of selected strains due to reduction in the frequency of LL-S birds (AL) relative to the total difference between selected and control strain (Ag) ranged between 4% and 14% for egg production and between 3% and 7% for egg weight. Thus the
TABLE 8. Examples of partitioning the differences between the performance of selected and control strains (Ag) into genetic gains (AQ) and gains due to reduction in the frequency of LLV shedders in the selected strains (A^) Based on 1977 data Strain
Strain type a Percentage of LLV shedders Hen-housed egg production 141-197 days: Average of LLV nonsheddersb Average of LLV shedders 0 Strain averaged Difference from control strain (Ag) Genetic gain (AQ) Gain due to reduction in the frequency of LLV shedders (A L ) Egg weight at 240 days (g): Average of LLV nonsheddersb Average of LLV shedders c Strain average Difference from control strain (Ag) Genetic gain ( A Q ) Gain due to reduction in the frequency of LLV shedders (A L )
18.8 222 197 217
51.5 49.8 51.2
7.7
283 258 281 64 61
4.3
247 222 246 29 25
57.2 55.5 57.1
57.7 56.0 57.6
5.9 5.7
6.4 6.2
18.3 224 199 219
53.7 52.0 53.4
.9 253 228 252 33 29
3.8 275 250 274 55 51
58.1 56.4 58.1
56.2 54.5 56.1
4.7 4.4
2.7 2.5 .2
C = Unselected control strain; S = strain selected for high egg production. For more details see Table 1. Estimates from Table 3 for strains 5, 1, 3, 7, and 8. Raw mean for strain 9. Reduction of hen-housed egg production by 25 eggs and of egg weight by 1.7 g in LL infected birds was assumed in all strains. Constructed from averages of LL positive and negative birds and their frequencies.
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
findings it can be deduced that the difference between the performance of a selected and a control strain can be partitioned to a component due to genetic gain and a component due to reduction in the frequency of the lower producing LL-S birds in the selected strain. The size of the latter part would depend on the magnitude of the LL effect on the trait under consideration and on the actual reduction in the proportion of LL-S birds in the selected strain compared with their proportion in the control strain. The partitioning of the gain resulting from selection can be thus expressed as:
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
2178
GAVORA ET AL. Marek's disease. Tech. Bull. No. 1412, ARS. USDA., Washington, DC. Spencer, J. L., L. B. Crittenden, B. R. Burmester, W. Okazaki, and R. L. Witter, 1977. Lymphoid leukosis, interrelations among virus infections in hens, eggs, embryos, and chicks. Avian Dis. 21:331-345. Spencer, J. L., L. B. Crittenden, B. R. Burmester, C. Romero, and R. L. Witter, 1976. Lymphoid leukosis viruses and gs antigen in unincubated chicken eggs. Avian Pathol. 5:221-226. Spencer, J. L., J. S. Gavora, and R. S. Gowe, 1978. Lymphoid leukosis virus infections in chickens under selection for high egg production. Jr. Amer. Vet. Med. Ass. 173:887. (Abstr.) Spencer, J. L., J. S. Gavora, and R. S. Gowe, 1979. Effect of selection for high egg production in chickens on shedding of lymphois leukosis virus and gs antigen into eggs. Poultry Sci. 58:279-284.
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
Brooksbank, and S. A. Hall, 1977. A survey of mortality in 51 caged laying flocks. Avian Pathol. 6:149-170. Sarma, P. S., H. C. Turner, and R. J. Heubuen, 1964. An avian leukosis group-specific complement fixation reaction. Application for the detection and assay of non-cytopathogenic leukosis viruses. Virology 2 3 : 3 1 3 - 3 2 1 . Schaeffer, L. R., and J. W. Wilton, 1976. Methods of sire evaluation for calving ease. J. Dairy Sci. 59:544-551. Sever, J. L., 1962. Application of a microtechnique to viral serological investigations. J. Immunol. 88:320-329. Sharma, T. M., 1979. Immunosuppressive effects of lymphoproliferative neoplasms of chickens. Avian Dis. 23:315-327. Siccardi, F. J., and B. R. Burmester, 1970. The differential diagnosis of lymphoid leukosis and
1980 Poultry Science 59:2165-2178 INTRODUCTION L y m p h o i d leukosis (LL) is a virus-induced lymphoblastic malignancy originating in t h e bursa of Fabricius. Chickens are infected either congenitally, t h r o u g h t h e egg (vertical transmission), or by direct c o n t a c t with infected chickens (horizontal transmission). Pathogenesis of L L has been described by Siccardi and Burmester ( 1 9 7 0 ) a n d b y C o u d e r t et al. ( 1 9 7 7 ) . Following infection of susceptible chickens, large a m o u n t s of virus are p r o d u c e d in m a n y tissues w i t h o u t any a p p a r e n t harmful effects.
1
Animal Research Institute Contribution No. 890. Animal Pathology Directorate, Health of Animals Branch, Agriculture Canada, Animal Diseases Research Institute, Nepean, Ontario, Canada K2H 8P9. 3 Animal Genetics Research Unit, USDA, SEA, AR, and Department of Animal Sciences, Purdue University, W. Lafayette, IN 47907. 2
Follicles of t h e bursa of Fabricius m a y b e c o m e t r a n s f o r m e d a n d remain quiescent until sexual m a t u r i t y of t h e chicken (16 t o 22 weeks) w h e n t h e cells start divisions, acquire neoplastic characteristics, and spread t o o t h e r visceral organs where t u m o r s m a y develop. Mortality usually does n o t occur until chickens have reached sexual m a t u r i t y . T w o levels of genetic resistance t o LL are recognized (Crittenden, 1 9 7 5 ) : 1) cellular resistance t o virus infection which is simply inherited b u t is very specific for t h e subgroup of virus, and 2) resistance t o t u m o r d e v e l o p m e n t in L L virus ( L L V ) infected birds which is m o r e c o m p l e x a n d appears t o be less subgroup specific. Mortality from L L is generally low. Purchase et al. ( 1 9 7 2 ) surveyed m o r e t h a n 1 4 5 , 0 0 0 birds over a period of a p p r o x i m a t e l y 18 m o n t h s . T h e y r e p o r t e d average L L m o r t a l i t y of 3.3 and 4.6% in Marek's disease vaccinated and nonvaccinated birds, respectively. Lifetime losses
2165
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
ABSTRACT Lymphoid leukosis (LL) is a virus-induced, lymphoblastic malignancy of chickens that can be congenitally transmitted. Mortality from LL is generally low. Effects of LL virus (LLV) on production and mortality were investigated in approximately 2000 Leghorn pullets in each of two consecutive years. The pullets were from nine strains developed in Ottawa, of which three were unselected control strains and six were strains under selection for up to 27 generations for high egg production and a complex of related commercially important traits. The overall frequency of birds shedding LL virus or gs antigen into eggs (LL-S) was significantly lower in the selected strains (3.9%) than in the control strains (18.5%), indicating that LLV may have negative effects on production and cause elimination of LL-S birds by selection. Such significant effects were indeed detected: the LL-S pullets produced to 497 days of age in 1976 and 1977, respectively, 30 and 25 eggs less per hen-housed than the nonshedders. The LL-S birds matured sexually later, produced smaller eggs at a lower rate, and their eggs had a lower specific gravity, indicating thinner shells. Mortality from all causes to 497 days was significantly higher in LL-S birds (+14.8%) in 1976. In 1977 the increase (+5.5%) did not reach statistical significance. In both years the mortality from LL itself remained very low. In another study, eggs from one of the control strains were incubated and hatched when the dams were 291 and 483 days old. The eggs from LL-S dams had 2.4% lower fertility and 12.4% lower hatchability. The effects on hatchability were more pronounced in the older dams. Since the lower production of LL-S birds results in a lower frequency of such birds in strains selected for high egg production, it is suggested that a part of the difference between the performance of the selected and control strains (Ag) is due to reduction in the frequency of LL-S birds ( A L ) rather than due to true genetic gain. In this study, the size of A L relative to Ag was estimated at 4 to 14% for egg production and 3 to 7% for egg weight. The negative effects of LLV infection on egg production, mortality, hatchability, and genetic gains show the desirability of producing chickens free of LLV infection. (Key words: lymphoid leukosis, egg production, mortality, selection, genetic gain)
2166
GAVORA ET AL.
MATERIALS AND METHODS
Strains of Chickens and Management of the Flocks. The strains tested for LLV infection have been previously described (Gowe, 1977; Gowe et al, 1973; Gavora et al, 1975) and the basic data on the strains are summarized in Table 1. Management of the flocks tested for LLV infection has been described by Gowe (1977), Gowe et al. (1973), and Spencer et al. (1979). Briefly, the chickens were reared in three-tiered group cages with sheet metal partitions between cages. Cages were randomly assigned to strains. Within each strain, day-old pullets from each sire family were randomly distributed through the cages. Approximately 75 chicks were placed in the top-tier cages on the day of hatch and were later divided among the three tiers. At 20 weeks, the pullets were distributed in a stratified random manner into individual 20 cm stairstep cages in two laying houses. The birds were fed mash rations ad libitum throughout the test. In both 1976 and 1977, all birds in the laying houses were divided into two groups that were fed slightly different diets: in 1976, one half of the birds (nutrition treatment 1 in Tables 3 to 5) received a diet with .4% phosphorus, while the ration fed to the other half (nutrition treatment 2 in Tables 3 to 5) had .35% of phosphorus. Similarly, in 1977, birds on nutrition treatment 1 (Tables 3 to 5) were fed a layer ration with 3.1% calcium, while birds on nutrition treatment 2 were given
free choice of oyster shell, spread on top of the same ration fed the other half of the birds. Individual egg production was recorded 5 days a week and then converted to a 7-day week basis. The birds were weighed at housing, at 1 year of age, and at the end of the egg production test (497 days). Egg weight and egg quality traits were measured on eggs laid within 5 consecutive days when the birds were approximately 240 and 450 days old. Collection of Egg Samples and Tests for LLV Infection. In 1976, eggs were collected from all hens of strains 5, 1, 4, 7, and 8 that were in one of the two laying houses between 255 and 259 days and between 283 and 287 days of age (Spencer et al, 1979). In 1977, eggs were similarly collected between 242 and 250 days of age from pullets of all strains in one of the two laying houses, but the size of the populations sampled was equalized to 220 pullets per strain. Results from LLV tests on one egg from each of 1935 and 1848 hens were used in the evaluation of LL effects on production in 1976 and 1977, respectively. To assess the consistency with which hens shed LLV into eggs, a second egg from each of 1394 hens was collected in 1977 when the hens were 468 to 476 days old. Detection of LLV and the group specific (gs) antigen in the albumen of unincubated eggs was reported by Spencer et al. (1976, 1977). In this study, the phenotypic mixing test (Okazaki et al, 1975) was used to detect LLV in the albumen of the eggs collected in 1976. The microcomplement fixation test (Sever, 1962; Sarma et al, 1964) was used to detect the gs antigen in the albumen of eggs collected in 1977. The gs antigen is an internal component of the virus common to all avian leukosissarcoma viruses. There is generally good agreement between results of tests for virus and gs viral antigen (Spencer et al, 1979). For the purposes of this presentation LLV shedders (LL-S) are birds that produced eggs in which LLV or gs antigen were detected. Examination of Effects of LLV Infection of the Dam on Fertility and Hatch ability. The hens of strain 5 tested for LLV infection in 1977 were pedigree mated by artificial insemination (1 male to 8 females) to a random sample of strain 5 males. Eggs were collected for incubation over two periods of two weeks each and chicks were hatched when the parents were 291 days old (hatch 1) and 483 days old (hatch 2). The eggs collected in the first week
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
from LL amounted to 1.2 to 3.4% in experimental strains of Leghorns and meat-type breeders maintained on the same premises as the birds in the current study (Gavora et al, 1975). In large field surveys conducted in Great Britain, 1.4% of total laying house mortality was attributed to LL in laying flocks (Randall et al, 1977) and 1.5% in broiler breeder flocks (Jones et al, 1978). Occasionally much higher losses from LL are experienced in some flocks (Crittenden and Witter, 1978). Evidence for negative effects of subclinical infections with LLV has been reported (Spencer et al, 1978; 1979; Chase, 1978). The main objective of the present study was to assess the economic losses from LL in terms of both mortality and effects of subclinical LLV infection on performance in egg production stocks. Consequences of such effects in strains selected for high egg production were also evaluated.
1959
Four commercial stocks crossed
Derived from strain 7
Derived from strain 7
Four commercial stocks crossed
C
s
s
c
7
9
8
10
1973
1969
1969
1969
c = unselected control strain; S = strain selected
a
1951
Seven Canadian R.O.P. unrelated stocks crossed
S
4
Derived from strain 4
S
2
1950
1971
Same as strain 5
S
3
1950
Derived from Strain 3
S
1
Narrow genetic base (Ottawa)
O
Year
Base population
Origin
5
Name Type
Sitrain
for high egg production.
None
None
478
1123
High hen-housed egg production
1130
475
Fertility, hatchability, viability, egg size, egg quality
None
High egg production Fertility, hatchability, rate viability, egg size, egg quality
None
1119
1128
High hen-housed egg production Fertility, hatchability, viability, egg size, egg quality
High egg production Fertility, hatchability, viability, egg size, rate egg quality
High hen-housed egg production 1123
476
1127
_
Fertility, hatchability, viability, egg size, egg quality
None
Secondary traits
Housed in laying house
High egg production Fertility, hatchability, rate viability, egg size, egg quality
None
Primary trait
Selection criteria
No. o
TABLE 1. Description of the strains, number of pullets housed, and incidence of shedding lymphoi
rom http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
GAVORA ET AL.
2168
u s
a
d
i
j:i
k:j:i "1
ah
e
lj:i
ijklm
= population mean = random effect of sire = fixed effect of LL status of dam within sire = random effect of dam within LL status within sire = fixed effect of hatch = interaction of LL status of a dam within sire with hatch = random individual effects
All-or-none data, such as mortality, fertility, or hatchability of an egg and classification of a newly hatched chick as normal or crippled were coded " 1 " for birds or embryos that died, for infertile eggs or eggs that did not hatch, and for chicks classified as crippled and " O " for the remaining eggs or birds. Harvey's (1972) LSMLMM program was used for the least squares analyses of the data. Linear model approach to the analysis of categorical data has been presented by Schaeffer and Wilton (1976) and by Grizzle et al. (1969) and was previously used for analysis of all-or-none data by Gavora etal. (1977).
Yijkim = u + a j + s j ; i + n k + b, + an i k RESULTS AND DISCUSSION
+ abil+nbkl+eijklm, where u a
S
i
j:i
n
k
b, an i k , ab n , and nbkl e
ijklm
= = = = =
population mean fixed effect of strain random effect of sire within strain fixed effect of nutrition treatment fixed effect of LL status of the individual
= interactions = random individual effects.
The statistical analyses of the strain 5 fertility anid hatchability data were performed
Yijklm =
u + s
i
+ a
j:i
+ d
k:j:i
+ h
l
+ ahjj.j + e i j k l m ,
where yijklm
=
f e r t m t y » hatchability of an individual egg or the classification of a chick as normal or crippled
Consistency of Shedding LLV into Eggs. Tests for shedding gs antigen into egg albumen were conducted on each of two eggs gathered several months apart from hens hatched in 1977. Only in 15 hens (1.1%) out of the 1,394 hens tested twice did the results of the second test differ from the results of the first test. In 5 instances, birds previously tested positive tested negative for gs antigen in the second test, while in 10 instances birds negative on the first test became positive on the second test. These results indicate 1) that gs antigen was shed into egg albumen with a high degree of consistency in the populations tested and 2) that the spread of LLV among adult hens (if any) did not increase the proportion of shedders significantly. Frequency of LL-S Birds. Strains under selection for high egg production had a significantly lower incidence of LL-S birds than unselected control strains (Table 2). Within the selected strains, breeders had been selected on the basis of their egg production and related traits but without paying any attention to their LLV status. In the control strains, breeders had been selected by a random process, completely disregarding both their egg production and LLV status. The lower frequency of LL-S in the
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
of each egg save period were stored in cryovac bags flushed with nitrogen to reduce the effects of egg storage on hatchability. All eggs were candled on day 18 of incubation and eggs with dead embryos and infertile eggs were identified and removed. On the day of hatch, all chicks hatched were examined and those with incompletely closed navel or with external malformations, as well as extremely weak chicks, were classified as "crippled". Statistical Analyses. The frequency of LL-S birds (Table 1) in strains 2 and 9 in 1977 was very low (1 and 2 LL-S birds in strains 2 and 9, respectively) and, therefore, these strains were excluded from the statistical analyses. The average number of pullets tested for shedding LLV into eggs per family of full sisters was 2.2 in 1976 and 1.4 in 1977. Thus, the infection status of the birds was often confounded with dam family, and therefore, the effect of dams was not included in the statistical analyses of the production and mortality traits. The following statistical model was used in the analyses:
LYMPHOID LEUKOSIS AND EGG PRODUCTION
2169
TABLE 2. Comparison of the frequency of birds shedding lymphoid leukosis virus (LLV) into eggs in selected and control strains and among breeders and the remaining birds that were not used in the reproduction of the strains 1976 Type of birds
Type of strains
Breeders
C*
Number tested 218
1977
Percent LLV Chishedders square 23.2
361 17.2 2.53
S
176 348
16.5 1.7
C
1193 394
5.4 20.0
S
1541
4.1
579
19.3
405 619
17.3 2.3
2180 984
4.7 18.5
2799
3.9
.67
.05 229 17.9 271 3.0
8.50'* Others All birds
Num- Percent ber LLV Chitested shedders square
.49 987 3.9 590 17.5
117.60**
7.30**
103.20** 1258
3.7
218.63**
C = Unselected control strains; S = strains selected for high egg production. For more details see Table 1. **P<.01.
selected than in the control strains (Table 2) suggests that production is negatively influenced by LLV infection. The lower production in LL-S birds then results in their elimination in the process of selection for high egg production. This is supported by the comparison of the frequency of LLV shedding among breeders and other birds not used in strain reproduction (Table 2). In the selected strains, the frequency of LL-S birds was significantly lower among breeders than among other birds in 1976 and also in the 1976 and 1977 data combined. However, there was no significant difference in the frequency of LLV positive birds among breeders and other birds within the control strains. These findings are consistent with the earlier conclusion (Spencer et al., 1979), based on the 1976 data, that strains under selection for high egg production have generally a lower incidence of LL-S birds than corresponding control strains. Of particular interest was control strain 10 (Table 1) synthesized in 1973 and maintained since unselected. The percentage of LL-S birds in this strain was in general agreement with results from the other two control strains indicating that, under the conditions of this study, the percentage of LL-S birds in populations maintained without selection tends to settle around 20%. It should be noted that the reduction in the percentage of LL-S birds in the selected strains cannot be attributed to an increase in genetic
cellular resistance to LLV infection. As shown by Spencer et al. (1979), the selected strains were at least as susceptible to LLV infection as were the control strains. The data on the percentage of LL-S birds are also in agreement with previous reports (Spencer et al., 1978, 1979; Chase, 1978) that LL-S birds tend to be eliminated in selection for high egg production. Comparison of Production and Mortality of LLV shedders and nonshedders. Tables 3 to 5 contain the estimated mean performance of LLV nonshedder birds in the selected and control strains tested, the effects of LLV shedding in terms of mean differences between LLV shedder and nonshedder birds, and degrees of freedom and mean squares from the analyses of variance. Large and significant effects of LLV shedding on hen-housed egg production were detected in both 1976 and 1977 (Table 3). It should be noted that, as mentioned above, the periods over which eggs were collected for tests of LLV shedding were from 255 to 287 days in 1976 and from 242 to 250 days in 1977. Thus, all mortality in this study was only from the time of the tests and consequently the hen-housed egg production from 141 to 273 days had little mortality component. The reduction of hen-housed egg production was largest towards the end of the 1-year egg production period (Table 3) and its size was quite consistent in the 2 years. No significant strain by LL status interaction
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
Others Breeders
Combined
Num- Percent ber LLV Chitested shedders square
C
7 8
6 334 1 1 6 6 1 1069
4 231 1 1 4 4 1 1688
Strain (St) Sire/strain Nutrition treatment (N) LLV Status (L) St X N StX L NX L Residual
13,100** 280** 229 2,164** 35 209 555 145
104/-3 105/-8**
94/-9** 110/-9** 911-1
9,321** 274** 4 1,433** 87 133 87 190
105/-6** 104/-4
3,117* 1,050** 83 15,113** 515 1,675 0 617
68/-15** 68/-15**
259A-27** 259/-33**
2,564** 832 1,541 13,075** 199 436 4,138* 717
**P«.01.
*P«.05.
Tests for LLV shedding were performed on eggs collected between 255 and 287 days in 1976 and between 292 in
37,895** 3,100** 1,005 58,938** 2,114 6,369 957 1,895
Mean squares
65/-22** 62/-7
66/-10
231/-18 274/-49**
58/-9 71/-25*
58/-4 73/-24**
93/-6 115/-10
270/-54*
60/-20*
114/-7
115/-7* 67/-26**
66/-21** 76/-1
88/-5 114/-2 111/ 3
88/-1 114/-4
236/-33** 281/ 4
259/-30**
64/-14** 57/-13 77/-20* 59/-5
68/-15**
105/-5**
104/-6**
C = Unselected control strain, S = strain selected for high egg production. For more details see Table 1.
1977
1976
Degrees of freedom
1 2
10
S
4
s c
C S S
5 1 3
a
of variation
Nutrition treatment
Strain
All birds
141-49
1976
1977
386-497 days 1976
1977
1976
141-273 days
Hen-housed egg production
Mean of LLV nonshedders/Difference LLV shedders
TABLE 3. Least squares mean hen-housed egg production and mortality for LLV nonshedders and LL effect expres nonshedders; degrees of freedom and mean squares for all birds tested for shedding lymphoid leuko
d from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
C S
c
7 8
10
4 228 1 1 4 4 1 1564
Strain (St) Sire/strain Nutrition treatment (N) LLV status (L) StX N StX L NX L Residual
6 330 1 1 6 6 1 1015
1977
160.5/ 6.9* 147.7/ 5.6
163.2/ 1.9 150.9/ 5.9**
7,203** 201** 97 436** 35 144 251 74
157.1/ 1.2 156.4/ 4.7**
149.1/
150.2/-1.8
5,073** 163** 286 601** 56 181* 96 86
154.8/ 2.0 155.6/ 4.4*
161.1/ 8.9**
1.2
168.1/ 4.0 151.4/-1.7 148.5/-2.3
155.2/ 3.2**
168.9/ 4.8* 150.4/ 4.2
156.7/ 3.0*
1,194** 184** 41 863** 94 528** 33 107
**P«.01.
*P«.05.
75. 73.
73.
69. 78.
76.
67.4 83. 73.
74.
197
1 3 1
1,9
1,1 2
Mean squares
76.8/ - 4 . 8 * 76.9/ - 3 . 5
69.5/ 1.4 78.7/-11.7**
78.3/ - 7 . 7 * *
74.3/ - 7 . 6 * * 83.6/ 4.8
76.9/ - 4 . 2 * *
1976
Hen-day egg product from first egg to 497 day:s (%)
C = Unselected control strain; S = strain selected for high egg production. For more details see Table 1.
1976
Degrees of freedom
S
4
1 2
Ca S S
5 1 3
Source of variation
Nutrition treatment
Strain
All birds
1977
1976
Age at the first egg (days)
Mean of LLV nonshedders/Difference LLV sh
TABLE 4. Least squares mean age at sexual maturity, hen-day egg production, and survivors egg production for difference between shedders and nonsbedders; degrees of freedom and mean squares for birds that (497 days) and were tested for shedding lymphoid leukosis virus (LLV) into
from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
4 6 228 330 (224)b (324)b 1 1 1 1 6 4 6 4 1564 1015 b (1409) b (825)
Strain (St) Sire/strain
25 90** 5 12 5 10
567** 25** 71* 101** 3 23 34** 11
337** 19**
55.9/ - . 6 55.6/-2.0**
.9
120** 50 13 16 87** 17
329** 46**
63.8/ .1 63.6/-2.3*
-.5
2.2
19
90*
178** 134** 22 31
141** 34**
Degrees of freedom in the analyses of the 450-day data.
**P<.01.
*P«.05.
0 122** 14 8 3 14
186** 37**
Mean squares
63.5/ - . 4 62.9/-3.1**
862/859/-
869/869/-
64.0/-2.0
829/-
61.0/-1.0 62.8/-1.9
862/ 874/
860/-
1976
64.3/-3.4
65.8A-2.9
65.4/-3.1*
59.0/
63.2/-1.7**
C = Unselected control strain; S = strain selected for high egg production. For more details see Table 1.
1977
1976
of variation
Degrees of freedom
56.4/-1.1* 56.1/ 1.6**
56.6/-1.6
62.2/ 65.4/
10
64.8/
53.7/ - . 6 56.2/ .8
54.6/-1.3 58.0/ - . 9
.2
57.6/
7 8
Nutrition treatment
Nutrition treatment (N) LLV status (L) StX N St X L N XL Residual
63.7/-1.1
1977
at 450 days
61.5/-2.1 64.7/-2.9
4
s s
55.8/-1.3**
56.2/-1.3** 51.5/ - . 4 57.2/ - . 1 57.7/-2.6* 51.4/-4.6**
Ca
1976
53.5/-2.5** 57.4/-2.1
5 1 3
1 2
Strains
All birds
1977
Egg weight (g)
1976
at 240 days
Mean of LLV nonshedders/Difference LLV shedd
TABLE 5. Least squares mean egg weight and egg specific gravity at 240 days and 450 days of age for LLV nonshe between shedders and nonshedders; degrees of freedom and mean squares for birds that survived to th were tested for lymphoid leukosis virus (LLV) shedding into eggs
from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
LYMPHOID LEUKOSIS AND EGG PRODUCTION
were observed in eggs from both younger and older birds. Body weights (not shown in the tables) of the LL-S birds tended to be slightly lower; differences, ranging between 30 and 60 g per bird, were not significant. Also, differences in Haugh units and in the percent of eggs with blood spots (not shown in the tables) were generally small and were mostly not significant. Differential responses of strains to LLV infection were indicated by significant strain by LLV status interactions in hen day percent and survivors' egg production in 1976 and in sexual maturity and 240 day egg weight in 1977. Significant nutrition treatment effects were found for egg weight and egg specific gravity. Significant nutrition treatment X LLV status interactions were found only in 1977 hen-housed egg production. The nutrition treatments will be dealt with further in a separate report, where the performance of all the birds, not just those that were tested for LLV shedding, will be evaluated. Table 6 shows the major causes of laying house mortality. An increase in mortality from all causes among the LL-S birds is indicated by the data. No deaths were attributed to LL in 1976; in 1977, mortality from LL was very low. The . 1 % mortality attributed to LL in the 1977 LL-S birds can be explained by the fact that in this study LL-S birds were identified by tests on egg albumen and, thus, some birds infected with LLV but which did not shed the virus into eggs may have been classified as LLV nonshedders. The total mortality in Table 6 is slightly different from that in Table 3. This is because Table 6 contains raw percentages while
TABLE 6. Association of increased laying bouse mortality percent with shedding lymphoid leukosis virus (LLV) into eggs Lymphoid leukosis status a
Cause of death
1976 LLV nonLLV shedders shedders
LLV nonshedders
LLV shedders
Reproductive disorders Marek's disease Fatty liver syndrome Lymphoid leukosis Other causes Total mortality
1.7 1.2 .7 0 1.4 5.0
1.8 1.9 .6 .1 1.5 5.9
3.4 3.4 1.3 1.3 3.5 12.9
4.0 3.3 2.6 0 5.5 15.4
1977
Based on the detection of lymphoid leukosis virus or the gs antigen in eggs collected between 255 and 287 days in 1976 and between 242 and 250 days in 1977.
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
is indicated by the analyses. However, the data in Table 3 suggest that strain 3 may have a better ability to tolerate LLV infection than the other strains. In 1976, laying house mortality was significantly higher (+14.8%) in the LL-S birds (Table 3). In 1977, mortality among the LL-S was also higher, but the difference (+5.5%) did not reach significance (P>.05). The absence of significant strain by LLV status interaction suggests that with the possible exception of strain 3, the effect of LLV shedding on the bird's survival was similar in all genotypes (strains) tested. As mentioned earlier, only mortality from the time of the LLV shedding tests was considered in this study. Since eggs for such tests were collected no earlier than at 242 days, the data in Table 3 do not include either early laying house mortality or rearing mortality. Hence, the effects of LL on mortality may be underestimated. As shown in Table 4, LLV shedding was associated with delayed sexual maturity. The mean effect was a delay of approximately 3 days, but the size of the effect varied widely among the individual strains. Egg production rate and the number of eggs produced by the birds that survived to 497 days was also lower for the LL-S birds, indicating that infections that caused LL-V shedding but did not result in the death of the birds were associated with reduced egg production. LL-S birds produced smaller eggs (Table 5) and their eggs had a lower average specific gravity, indicating thinner egg shells which could lead to higher egg breakage. Such effects
2173
2174
GAVORA ET AL.
physiological processes. These effects ranged from susceptibility to infectious diseases (Marek's disease) through deaths from reproductive disorders and fatty liver syndrome to effects on fertility, hatchability, sexual maturity, egg production rate, egg weight, and egg shell strength. A majority of the effects was associated with subclinical infection with LLV because mortality from LL was very low (Table 6). Although the association of LL-S status with lower production and higher mortality was clearly demonstrated in this study, the possibility that the association was not due to a direct effect of LLV cannot be eliminated. If, for example, shedding of LLV into eggs occurred primarily among hens less competent immunologically, then their low production and high mortality may have been the result of the lower immunological competence rather than the effect of LLV. Mechanisms through which the subclinically infected birds may be affected by LLV have not been investigated in this study. However, it is noteworthy that in electron micrographic studies DiStefano and Dougherty (1968, 1969) found LLV in the ovary and testis, pituitary, thyroid, parathyroid glands, and in the adrenal cortex of LLV infected chickens that appeared healthy. This suggests possible effects of LLV on a wide range of endocrine functions and may at least partly explain the negative effects on economic traits observed in this study. Research on the immunosupressive effects of LL was recently reviewed by Sharma (1979). He concluded that most researchers noted some depression in antibody synthesis by LLVinfected chickens. Although the exact mechanism of this impairment is unknown, it may be related to the affinity of LLV to the bursa of Fabricius. In vitro studies indicated that LLV infection also lowers cellular immune competence. It appears that immune response may be influenced by LLV infection even without the accompanying cellular transformation. Thus, immunosupression associated with LLV infection may explain the increased mortality from causes other than LL observed among the LL-S pullets in this study. Effects of LL on Selection Gains. As discussed earlier, several aspects of production were negatively influenced by LLV shedding. It has also been demonstrated that the frequency of LL-S birds is reduced in strains under selection for high egg production. From these
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
least squares means of total mortality are shown in Table 3. Effects of LLV Infection on Fertility and Hatchability. Mean fertility, early and late embryonic mortality, hatchability, and percentage of crippled chicks for LLV nonshedder dams, as well as mean differences in these parameters between the LLV shedder and nonshedder dams, are shown in Table 7, together with degrees of freedom and mean squares. Significant decrease in fertility, hatchability from fertile eggs, and consequently in hatchability from eggs set was associated with LLV shedding by the dam. The decrease in hatchability of fertile eggs was primarily due to higher mortality of embryos prior to day 19 of incubation. Since the eggs classified as infertile by candling were not broken and examined for presence of zygotes, it is possible that some of the reduction in fertility might have been due to early embryonic mortality not distinguishable by candling. The increase, by 1.3%, in the percentage of crippled chicks among the progeny of LL-S dams, although substantial relative to the 1.9% crippled chicks observed in the progeny of negative dams, did not reach statistical significance. Both hatchability and percentage of crippled chicks were affected by hatch. Since the insemination, egg save, and incubation procedures were similar in both hatches, hatch effects were likely primarily associated with age of the brids. Thus the observation of more severe effects of LLV infection on hatchability and percentage of crippled chicks in Hatch 2 (when the dams were 487 days old; Table 7) coincides with the observation that LLV infection had the largest effect on egg production in the last part of the production period (386 to 497 days; Table 3). Although in the study on hatchability only one strain of chickens was used, the size of the effect of LLV status of the dam on production of day-old chicks (hatchability of eggs set) is large enough to be of concern and to warrant more investigation. In selected strains, the reduced hatchability could also be expected to reduce the contribution of progeny from LL-S dams that have not been eliminated by selection. The above data on the association of LLV shedding with lower production and higher mortality from all causes suggest the effects of LLV on a number of distinctly different
Degrees of freedom
1 2
2,000
3,063** 1,271** 10,871** 2,148** 529
20
20 105 1 14 1732
1,683** 780 150 525 511
Mean squares
6.2/ 5.4 8.9/ 17.1
20 103 1 14 1611
20
Degrees of freedom
7.6/ 11.2**
33,533**
Mean squares
Degrees of freedom
89.3/- -2.4**
90.2/- -1.1 88.5/- -3.9
**P«.01.
1,692** 757** 3,134** 1,815** 347
1,769
Mean squares
3.1/ 9.0 14.6/-3.6
10.2/ 2.8**
19-21 days
Embryonic mortality (%)
1-18 days
Significance of the effect of the dam's LL status was not tested within hatch.
Sire (S) 20 LLV status of dam (Ds)/S 21 Dam (D)/Ds/S 109 Hatch (H) 1 H X Ds/S 14 Residual 1947
Source of variation
Hatch*
All dams
Fertility (%)
20 105 1 14 1732
20
Degrees of freedom i
6,322** 2,654** 41,862** 57,905** 990
5,439
Mean squares
83.9/ - 5 . 6 78.5/-19.2
81.2/-12.4**
fertile eggs
Hatcha
Mean of LLV nonshedder dams/Difference LLV shedder dams
TABLE 7. Least squares mean fertility, embryonic mortality, hatchability, and percentage of crippled chicks for as difference between shedder and nonshedder dams; degrees of freedom and mean squares from
from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
2176
GAVORA ET AL.
where Ag is the difference in performance between a selected and a control strain, AQ'IS the genetic gain, and AL is the gain due to reduction in the frequency of LL-S birds resulting from selection. These two parts can be
estimated as: AG = Xi -x2, and A
L
=
q2(x2-y2)-qi(xi-yi),
where ~k~t and ~x2 are mean performances of the birds that did not shed LLV in the selected and control strains, respectively, y"t and y 2 are the respective mean performances, and q ! and q 2 are the respective frequencies of LL-S birds in the selected and control strains. Table 8 shows examples of Ag, AQ, and AL calculated for egg production and egg weight for control strain 5 and corresponding selected strains 1 and 3, and for control strain 7 and corresponding selected strains 9 and 8. The calculations were based on 1977 data. The size of the improvement in production of selected strains due to reduction in the frequency of LL-S birds (AL) relative to the total difference between selected and control strain (Ag) ranged between 4% and 14% for egg production and between 3% and 7% for egg weight. Thus the
TABLE 8. Examples of partitioning the differences between the performance of selected and control strains (Ag) into genetic gains (AQ) and gains due to reduction in the frequency of LLV shedders in the selected strains (A^) Based on 1977 data Strain
Strain type a Percentage of LLV shedders Hen-housed egg production 141-197 days: Average of LLV nonsheddersb Average of LLV shedders 0 Strain averaged Difference from control strain (Ag) Genetic gain (AQ) Gain due to reduction in the frequency of LLV shedders (A L ) Egg weight at 240 days (g): Average of LLV nonsheddersb Average of LLV shedders c Strain average Difference from control strain (Ag) Genetic gain ( A Q ) Gain due to reduction in the frequency of LLV shedders (A L )
18.8 222 197 217
51.5 49.8 51.2
7.7
283 258 281 64 61
4.3
247 222 246 29 25
57.2 55.5 57.1
57.7 56.0 57.6
5.9 5.7
6.4 6.2
18.3 224 199 219
53.7 52.0 53.4
.9 253 228 252 33 29
3.8 275 250 274 55 51
58.1 56.4 58.1
56.2 54.5 56.1
4.7 4.4
2.7 2.5 .2
C = Unselected control strain; S = strain selected for high egg production. For more details see Table 1. Estimates from Table 3 for strains 5, 1, 3, 7, and 8. Raw mean for strain 9. Reduction of hen-housed egg production by 25 eggs and of egg weight by 1.7 g in LL infected birds was assumed in all strains. Constructed from averages of LL positive and negative birds and their frequencies.
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
findings it can be deduced that the difference between the performance of a selected and a control strain can be partitioned to a component due to genetic gain and a component due to reduction in the frequency of the lower producing LL-S birds in the selected strain. The size of the latter part would depend on the magnitude of the LL effect on the trait under consideration and on the actual reduction in the proportion of LL-S birds in the selected strain compared with their proportion in the control strain. The partitioning of the gain resulting from selection can be thus expressed as:
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
2178
GAVORA ET AL. Marek's disease. Tech. Bull. No. 1412, ARS. USDA., Washington, DC. Spencer, J. L., L. B. Crittenden, B. R. Burmester, W. Okazaki, and R. L. Witter, 1977. Lymphoid leukosis, interrelations among virus infections in hens, eggs, embryos, and chicks. Avian Dis. 21:331-345. Spencer, J. L., L. B. Crittenden, B. R. Burmester, C. Romero, and R. L. Witter, 1976. Lymphoid leukosis viruses and gs antigen in unincubated chicken eggs. Avian Pathol. 5:221-226. Spencer, J. L., J. S. Gavora, and R. S. Gowe, 1978. Lymphoid leukosis virus infections in chickens under selection for high egg production. Jr. Amer. Vet. Med. Ass. 173:887. (Abstr.) Spencer, J. L., J. S. Gavora, and R. S. Gowe, 1979. Effect of selection for high egg production in chickens on shedding of lymphois leukosis virus and gs antigen into eggs. Poultry Sci. 58:279-284.
Downloaded from http://ps.oxfordjournals.org/ at University of Cambridge on May 24, 2015
Brooksbank, and S. A. Hall, 1977. A survey of mortality in 51 caged laying flocks. Avian Pathol. 6:149-170. Sarma, P. S., H. C. Turner, and R. J. Heubuen, 1964. An avian leukosis group-specific complement fixation reaction. Application for the detection and assay of non-cytopathogenic leukosis viruses. Virology 2 3 : 3 1 3 - 3 2 1 . Schaeffer, L. R., and J. W. Wilton, 1976. Methods of sire evaluation for calving ease. J. Dairy Sci. 59:544-551. Sever, J. L., 1962. Application of a microtechnique to viral serological investigations. J. Immunol. 88:320-329. Sharma, T. M., 1979. Immunosuppressive effects of lymphoproliferative neoplasms of chickens. Avian Dis. 23:315-327. Siccardi, F. J., and B. R. Burmester, 1970. The differential diagnosis of lymphoid leukosis and