J. Anim Sci. 2008. 86:1737-1746. doi:10.2527/jas.2007-0650
© 2008 American Society of Animal Science
Association of the prion protein gene with individual tissue weights in Scottish Blackface sheep1
R. M. Sawalha*,2,
S. Brotherstone
,
N. R. Lambe* and
B. Villanueva*
* Scottish Agricultural College, West Mains Road, Edinburgh EH9 3JG, UK; and
School of Biological Sciences, University of Edinburgh, West Mains Road, Edinburgh EH9 3JT, UK
 |
Abstract
|
|---|
This study investigated associations of prion protein (PrP) genotype with body composition and weight traits of Scottish Blackface ewes. Body composition was predicted using computer tomography (CT) scans to estimate muscle, carcass fat, internal fat, and bone weights. The traits were measured at 4 key seasonal production points (pre-mating, pregnancy, midlactation, and weaning) over 4 production cycles (2 to 5 yr old). There were 2,413 records for each of the CT traits measured on 335 ewes, and 26,649 records for each of the body condition score and BW traits for 2,356 ewes. From 1999 to 2004, animals were genotyped to determine polymorphisms at codons 136, 154, and 171, which are associated with scrapie susceptibility. Four alleles were found in the population (ARR, AHQ, ARQ, and VRQ). The data were analyzed using a linear mixed random regression model assuming that the direct additive genetic effect was a 2nd order Legendre polynomial function of time. The PrP genotype was included in the model as a fixed effect along with other fixed factors with significant effects (P < 0.05). Five separate analyses were carried out for each trait, depending on the method of classifying the PrP genotype. In the first analysis, animals were categorized according to the genotype. Only the 5 most common genotypes (ARR/ARR, ARR/AHQ, ARR/ARQ, AHQ/ARQ, and ARQ/ARQ) were included. In the last 4 analyses, animals were categorized according to the number of each PrP allele carried. For CT traits and body condition score, results showed that the PrP genotype has no association with the overall mean of the traits (averaged over age). For BW, ewes without the ARQ allele were at least 0.5 kg heavier than ARQ homozygous and heterozygous ewes. On the other hand, there was a significant interaction between PrP genotype and age of the ewe (i.e., the effect of PrP genotype was not the same at different ages for 5 out of the 6 traits studied). In general, ARQ carrying ewes mobilized more fat reserves at times of nutrient deficiency, such as during lactation, and gained it back more quickly by the mating season (when nutrients became abundant) than non-ARQ carriers. Therefore, selecting against this allele would have consequences on BW and seasonal mobilization of body reserves. The number of VRQ alleles (the most scrapie susceptible allele) carried was not significantly associated with any of the traits.
Key Words: computer tomography • growth • prion protein gene • PrP gene association • scrapie • sheep
|
INTRODUCTION
|
|---|
Susceptibility to classical scrapie in sheep is strongly associated with polymorphisms of the prion protein (PrP) gene (Goldmann et al., 1994
). Five alleles (ARR, ARQ, AHQ, ARH, and VRQ) are commonly found in sheep populations. The ARR allele is known to confer the greatest resistance, whereas the VRQ allele, when present, is associated with the highest risk of the disease. The persistence and segregation of PrP alleles linked with susceptibility to scrapie for over 2 centuries since the discovery of the disease may suggest that the gene is associated with fitness or performance traits under selection. In fact, postnatal lamb survival was found to be significantly associated with PrP genotype (Sawalha et al., 2007a). Most association studies investigated lamb growth traits (e.g., Vitezica et al., 2005; Isler et al., 2006). However, no study has been published about possible PrP gene association with seasonal changes of body and tissue weights.
Ewes raised under extensive production systems (e.g., hill sheep), have been shown to undergo sub-stantial seasonal fluctuations in body fat and muscle weights during the annual production cycle (Lambe et al., 2004). Efficient tissue mobilization is essential to meet the nutrient requirements of reproduction and lactation while ensuring survival in response to fluctuations in feed availability and other environmental conditions (Russel et al., 1968; Lambe et al., 2005). Computer tomography (CT) measurements provide accurate predictions of weights of different types of tissues of live animals (Lambe et al., 2003b). Multiple CT measurements allow for accurate modeling of the traits over time using advanced statistical techniques such as random regression.
The objective of this study was to investigate possible associations of PrP genotype with individual CT muscle, fat, bone, and BW and condition score in Scottish Blackface sheep (the main hill sheep breed in the United Kingdom) over multiple production cycles.
|
MATERIALS AND METHODS
|
|---|
All procedures involving animals were governed by an animal ethics committee at the Scottish Agricultural College and were performed under the United Kingdom Home Office license following the regulations of the Animals Act 1986.
Animals, Farms, and Management
Scottish Blackface ewes from 2 Scottish Agricultural College (SAC) research farms were involved in the study. The farms and management system have been described previously (Conington et al., 2001; Conington et al., 2006). In brief, Castlelaw farm is located in the Pentland hills near Edinburgh, whereas Kirkton farm is located near Crianlarich in the Western High-lands, with a harsher climate and poorer vegetation quality than Castlelaw farm. Both farms comprised approximately 600 ewes. The animals were raised under typical extensive management practices and conditions for Scottish hill sheep, with the exception that mating (where single-sire mating groups were used) and lambing were both performed in improved areas near farm buildings to allow pedi-gree recording. The grazing area was divided according to steepness and feed availability. Ewes were assigned to different areas and a record of the grazing area was kept for each ewe at all times. At each farm, ewes were naturally mated in about 16 single-sire groups per year, to enable pedigree recording. In addition, on both farms, 40 ewes per year were mated by artificial insemination to 2 rams (from the UK Blackface sire reference scheme) to provide genetic links with commercial flocks and between the 2 flocks (Simm et al., 2001). Production figures on the SAC farms tended to be greater than breed averages for similar environments (SAC, 2007), with weaning percentages (number of lambs weaned as a percentage of ewes mated) being 114 to 133% at Castlelaw and 84 to 121% at Kirkton farm for the period 1998 to 2006.
Data and Traits
The traits studied were CT carcass muscle, CT carcass fat (subcutaneous and inter-muscular fat), CT internal body cavity fat, CT bone, body condition score, and BW over 4 annual production cycles of the ewes. The ewes were recorded at 4 focal production points each year: mating (mostly in November, at an average of 15 d before mating), pregnancy (in April, at an average of 41 d before lambing), lactation (mostly in June, at an average of 51 d after lambing), and weaning (in August, at an average of 115 d after lambing). The first measurement was taken at the first mating of the yearlings (as early as 18 mo of age) and the last was recorded at weaning time for ewes up to 5 yr old. There were 2,413 records for each of the 4 CT traits measured on 335 ewes from Castlelaw farm, and 26,649 for each of body condition score and BW records measured on 2,356 ewes from both Castlelaw and Kirkton farms. The number of records and averages of the traits by ewe age and event of recording are presented in Figure 1.
Cross sectional CT images were taken at 5 anatomical positions for each ewe: 8th thoracic vertebra, 2nd lumbar vertebra, 5th lumbar vertebra, hip, and ischium as recommended in a previous study (Lambe et al., 2003a). The CT data, in the form of areas and average densities at each cross sectional position, were used to estimate weights of the 4 types of tissues (muscle, car-cass fat, internal fat, and bone) for each ewe at each scanning event. Prediction equations developed in a previous study using dissection, CT, and live weight data of Scottish Blackface ewes were used (Lambe et al., 2003b). Body condition score was a subjective estimate of the depth of subcutaneous fat and the size of the loin muscle and was scored by palpation on a scale of 0 to 5 (Russel et al., 1969). A score of 5 indicates ewes with a full loin muscle and very thick fat cover, whereas a score of 0 indicates severely emaciated ewes with extremely thin loin muscle and no fat cover.
All animals present at the 2 farms from 1999 to 2004 were genotyped for codons 136, 154, and 171 of the PrP gene using proprietary technology (Orchid Cellmark Europe Ltd, Oxfordshire, UK). This technology allows the distinction among ARR, AHQ, ARQ, ARH, and VRQ alleles. Frequencies of PrP genotypes and alleles are presented in Table 1. Males carrying the VRQ allele were not used as sires at either of the 2 farms, and, therefore, there were no homozygous VRQ animals in the data set. As in the national UK Scottish Blackface population (Eglin et al., 2005), the ARH allele was not present in these flocks.
Statistical Analyses
The data were analyzed using random regression with a linear mixed model using the software ASReml (Gilmour et al., 2002). The average response curve for each trait was modeled as a fixed orthogonal polynomial of age at measurement of order 11 to 13, which is equivalent to fitting a cubic curve for each year of recording. A high order of polynomial has been shown to be required when modeling live weight by age using random regression (Meyer, 2000). The direct additive genetic and the permanent environmental effects were found to be significant and were included in the model as random effects. The additive genetic effects were modeled as quadratic Legendre polynomials to allow each animal to deviate from the average curve. Measurement error variance was found to be different by age. Therefore, estimates of residual variances were allowed to vary for each of the 4 recording events and the 4 ages, resulting in 16 estimates of residual error variance for each trait. Estimates of variance components were obtained and have been published previously (Lambe et al., 2004).
The model also included grazing area (13 levels for CT traits and 33 levels for body condition score and BW), farm (2 farms, but not for CT traits), and year of birth of the ewe (1996 to 2004) as fixed effects. The model also included the fixed effect of number of lambs born (0, 1, 2, or more) in the previous production cycle for traits recorded at mating. Traits recorded during pregnancy, lactation, and weaning were adjusted for number of lambs born in the same year as for the record itself by including litter size born as a fixed effect in the model. The possible associations of PrP genotype with performance traits were studied by including the genotype as a fixed effect in the model. The possible existence of an interaction of PrP genotype with age was also tested by including PrP genotype by orthogonal polynomial of age at measurement in the model. This allowed testing of the differences between means of different PrP genotypes at different ages.
Five different analyses differing in the PrP genotype classification were performed for each trait. The first analysis included only ewes with the 5 most common genotypes (ARR/ARR, ARR/AHQ, ARR/ARQ, AHQ/ ARQ, and ARQ/ARQ); i.e., rare genotypes (VRQ carriers and AHQ/AHQ animals) were excluded. The other 4 analyses were based on the number of each of the 4 PrP alleles carried. For example, based on the number of ARQ alleles carried (analysis 4), the animals were classified as ARQ/ARQ, ARQ/xxx, and xxx/xxx, where xxx represents any PrP allele other than the ARQ. A similar classification approach was applied for the other PrP alleles (ARR, AHQ, and VRQ). For testing the interaction of PrP genotype with age, only results from the first analysis will be presented.
The means of the traits by PrP genotype were predicted at 4 recording events during the 4 production cycles considered. The model predictions were marginal means across all the explanatory variables in the model except PrP genotypes, as described by Gilmour et al. (2004). Individual contrasts of model predictions were only performed if the overall test was significant (P < 0.05). The model predictions at each of the 16 points described above were compared for different PrP genotypes using Bonferroni corrected multiple range tests (Shaffer, 1995). For example, the correction made for the analysis that considered the 5 most common PrP genotypes was based on 10 multiple range tests. Another contrast testing the change in carcass fat weight by age for ARQ/ARQ and ARR/AHQ genotypes was carried out for illustrative purposes.
|
RESULTS
|
|---|
The number of available records decreased with age of the ewe (Figure 1
). The number of records within year was also variable and the largest number available was mostly at mating time. There was a distinct annual cyclic oscillation in the mean of individual traits. The standard deviation also varied with age and event, with a gradual increase by age. Generally, records taken at mating and pregnancy had lower variability than records measured during lactation and at weaning, except for BW records.
The first analyses comparing the overall trait means (the average over all ages and events) for the 5 most common genotypes showed no associations (P > 0.05; Table 2). Similarly, the number of ARR (the most scrapie resistant), AHQ or VRQ (the most scrapie susceptible) alleles carried showed no association (P > 0.05) with any of the traits studied. On the other hand, the number of ARQ alleles carried was associated (P = 0.03) with mean BW. Ewes without the ARQ allele were heavier by 0.67 ± 0.29 kg than ARQ homozy-gous ewes (P = 0.03) and by 0.56 ± 0.25 kg than ARQ heterozygous ewes (P = 0.04). However, BW means of ARQ homozygous and heterozygous ewes were not different (P = 0.47). The model predicted BW means for ARQ homozygous, heterozygous, and noncarrier ewes, averaged over ages were 56.74 ± 0.30, 56.85 ± 0.28, and 57.41 ± 0.34 kg, respectively. The number of ARQ alleles carried was not associated (P > 0.05) with the mean of any other trait besides BW (Table 2).
Figures 2, 3, 4, 5, 6, and 7 show the model predicted means by age and recording event for the 5 most common PrP genotypes for CT muscle, carcass fat, internal fat, and bone weights, and for body condition score and BW, respectively. There was statistically significant (P < 0.05) interaction between PrP genotype and age for all traits studied except CT muscle weight (results not shown), suggesting the effect of the PrP genotype was different at the different ewe ages. Ewes with the ARR/ ARR and ARR/AHQ genotypes tended to have greater means for CT tissue and BW at several recording events and ages compared with ewes with the other genotypes. The mean CT muscle weight at different ages and recording events was not associated (P > 0.05) with the PrP genotype, though the differences between means by PrP genotype were in the same direction as for the other significantly affected traits (Figure 2). Particularly at older ages, ewes having the ARQ allele lost more carcass and internal fat weight at times of nutrient deficiency (by middle of lactation) and put on more weight when nutrients became abundant (by mating season, Figures 3 and 4). For example, when comparing ARQ/ ARQ with ARR/AHQ ewes, the first group (ARQ/ARQ) lost 2.22 + 0.12 kg of carcass fat from mating to middle of lactation when they were 4 yrs old and gained 2.65 + 0.12 kg by the next mating season, while the second group (ARR/AHQ) lost 1.22 + 0.12 kg and gained 2.01 + 0.12 kg during the same period. These differences between the genotypes considered (ARQ/ARQ and ARR/ AHQ) were statistically significant (P < 0.05).
View larger version (27K):
[in this window]
[in a new window]
|
Figure 2. Model predicted means of computer tomography (CT) muscle weight by PrP genotype and age. Within age and event, prion protein (PrP) genotype by age was not significant (P > 0.05). Animals were recorded 4 times per year (M = mating, P = pregnancy, L = lactation, and W = weaning) for 4 yr (from 2 to 5 yr of age).
|
|
View larger version (32K):
[in this window]
[in a new window]
|
Figure 3. Model predicted means of computer tomography (CT) carcass fat weight by prion protein (PrP) genotype and age. a–dWithin age and event, points with different letters differ (P < 0.05). Animals were recorded 4 times per year (M = mating, P = pregnancy, L = lactation, and W = weaning) for 4 yr (from 2 to 5 yr of age). Oval symbols highlight the specific genotypes or group of genotypes that differ significantly.
|
|
View larger version (29K):
[in this window]
[in a new window]
|
Figure 4. Model predicted means of computer tomography (CT) internal fat weight by prion protein (PrP) genotype and age. a,bWithin age and event, points with different letters differ (P < 0.05). Animals were recorded 4 times per year (M = mating, P = pregnancy, L = lactation, and W = weaning) for 4 yr (from 2 to 5 yr of age). Oval symbols highlight the specific genotypes or group of genotypes that differ significantly.
|
|
View larger version (26K):
[in this window]
[in a new window]
|
Figure 5. Model predicted means of computer tomography (CT) bone weight by prion protein (PrP) genotype and age. a,bWithin age and event, points with different letters differ (P < 0.05). Animals were recorded 4 times per year (M = mating, P = pregnancy, L = lactation, and W = weaning) for 4 yr (from 2 to 5 yr of age). Oval symbols highlight the specific genotypes or group of genotypes that differ significantly.
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Figure 6. Model predicted means of body condition score by prion protein (PrP) genotype and age. a,bWithin age and event, points with different letters differ (P < 0.05). Animals were recorded 4 times per year (M = mating, P = pregnancy, L = lactation, and W = weaning) for 4 yr (from 2 to 5 yr of age). Oval symbols highlight the specific genotypes or group of genotypes that differ significantly.
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Figure 7. Model predicted means of BW by PrP genotype and age. a,bWithin age and event, points with different letters differ (P < 0.05). Animals were recorded 4 times per year (M = mating, P = pregnancy, L = lactation, and W = weaning) for 4 yr (from 2 to 5 yr of age). Oval symbols highlight the specific genotypes or group of genotypes that differ significantly.
|
|
The mean of CT carcass fat was associated (P 
0.02) with PrP genotype at pregnancy when the ewes were 2 to 3 yr old and at all scanning events when the ewes were 4 to 5 yr old (Figure 3). Generally, ewes with the ARR/ARR or ARR/AHQ genotype had greater (P 0.02) CT carcass fat than ewes with the other genotypes when they were 4 to 5 yr old. Similarly, internal fat was greater (P = 0.04) for ewes with the ARR/ARR genotype than for ewes with the AHQ/ARQ or ARQ/ ARQ genotype at weaning when the ewes were 4 yr old (Figure 4). In the same way, ewes with the ARR/ARR genotype had greater (P 0.04) mean bone weight than AHQ/ARQ ewes at mating and pregnancy when they were 4 yr old (Figure 5). The ARR/ARR genotype was associated with greater (P 0.03) body condition score than the other genotypes at weaning time when the ewes were 2 yr old, whereas the ARQ/ARQ genotype was associated with a greater (P = 0.01) score when the ewes were 5 yr old at weaning time (Figure 6). As with CT traits, the mean BW of ewes with the ARR/ARR or ARR/AHQ genotypes was greater (P 0.04) than that of ewes with the other PrP genotypes at several scanning events and ages (Figure 7).
|
DISCUSSION
|
|---|
In this study we investigated possible associations of the PrP genotype with 4 CT tissue weight traits, body condition, and BW of ewes at 2 to 5 yr of age. Except for CT muscle, all 5 other traits studied were affected (P < 0.05) by PrP genotype differently at different ewe ages (i.e., the differences between the means by different PrP genotypes were not the same at different ages). The most prominent estimate of this interaction effect was for carcass fat weight, where the differences between means for different PrP genotypes were different by age (P < 0.05). There was general agreement between the pattern of depletion and repletion for most body tissues over several production cycles by PrP genotype. For example, tissue mobilization, as measured by the mean CT weight of muscle, carcass fat, and internal fat by age and PrP genotype was very similar (Figures 2 to 5). However, the magnitude of fat weight (both carcass and internal) depletion and repletion was greatest for ewes with the ARQ allele. As Lambe et al. (2007) found that ewes with the greatest amounts of annual tissue mobilization rear heavier lambs, our results may suggest that ARQ carrying ewes could outperform others in this respect.
Several other studies have investigated possible associations of different sheep performance traits with PrP genotype. The traits studied have included lamb growth (Vitezica et al., 2005; Isler et al., 2006; Sawalha et al., 2007b), ewe reproductive traits (Ponz et al., 2006; Vitezica et al., 2006; Casellas et al., 2007), and dairy traits (de Vries et al., 2005; Álvarez et al., 2006). The current study is the first to investigate the possible associations with seasonal tissue mobilization and BW of ewes, which are likely to affect ewe survival and reproductive performance. de Vries et al. (2004, 2005, 2006) investigated associations of PrP genotype with body size measurements (height, length, and girth), back muscle and fat depth, and subjective type traits (conformation and muscle mass) in several breeds at about 212 d. Results included a significant association of the ARR allele with measures of size. Their results indicated that in East Friesian sheep ARR carrier animals were smaller, and in a German black-headed mutton breed ARR carriers had less back muscle than noncarriers. In our study the ARR allele was not associated (P > 0.05) with the overall mean averaged over age of any of the traits studied. However, ARR homozygous ewes tended to be heavier than others at some ages and scanning events.
The availability of large numbers of multiple records, made possible by the use of CT scanning in our study, may have helped in revealing the rather small but statistically significant associations found. Rams were mostly used for only a single breeding season, which may rule out attributing such associations to a founder effect. Selective genotyping was also not a factor as all animals in the 2 flocks were genotyped (de Vries et al., 2005, 2006; Vitezica et al., 2005). Furthermore, breeding animals were not selected on PrP genotype, except for avoiding the use of VRQ carrier rams.
Random regression models do not require the assumption of constant variances or correlations of traits with multiple records over time compared with univariate fixed regression models, and they may require estimation of fewer parameters than multivariate models (Meyer and Hill, 1997; Meyer, 2000). Random regression models have advantages over fixed curve models as individual additive genetic curves for each animal can be estimated. Additionally, the correlation is lower between estimates of the random coefficients when using orthogonal polynomials compared with the estimates of fixed parametric curves (Jamrozik et al., 1997; Lewis and Brotherstone, 2002). The means of the records at different ages and scanning events were highly variable for all traits in this study, and, therefore, high order polynomials (11 to 13) were used to model the trend curves. Random regression models also allow for inter-polation of a trait between the points with records. This is of particular importance when dealing with missing or incomplete data, which is often the case when using data taken over extended periods as in this study. The variability of the records was not the same at different ages and scanning events, and this was accounted for by using heterogeneous measurement error variance.
Mobilization of body reserve depots can occur merely in response to feed availability and weather conditions. This has been shown in both deer and sheep where seasonal fluctuations in fat and muscle reserves were reported in non-reproducing females (Weber and Thompson, 1998; Lambe et al., 2004). To investigate if our results were in response to nutritional demand as influenced by litter size, the data of all traits were re-analyzed excluding the number of lambs born from the model. Results from these later analyses were in very close agreement with the results after adjusting for litter size at birth, which we suggest indicates that the presence or absence of association of PrP genotype with ewe BW traits was not related to ewe reproductive performance.
Tissue mobilization is essential to meet the variable nutritional requirements at different seasonal production phases within each annual production cycle. Body reserves in the forms of fat (carcass and internal) and, to a lesser extent, muscle, have been shown to be depleted in times of negative energy balance, such as during early to mid lactation, and repleted in times of sur-plus energy, for example during late lactation and the dry period (Lambe et al., 2003a). Nutrient availability, particularly under extensive hill sheep production systems, is highly variable. Both lamb and ewe survival may be related to the potential of the ewe to store nutrients when feed is abundant, and to use it at times of scarcity (winter) or when demand is high (late pregnancy and early lactation). The efficiency of tissue mobilization is a major factor in the sustainability of hill sheep production systems. Therefore, breeding goals for hill sheep need to aim not only at productivity, but also toward more sustainable production (Lambe et al., 2004).
We also investigated the possible association of PrP genotype with ultrasonic backfat and muscle depths using data collected at mating time for ewes from 2 to 5 yr of age and found no evidence of association (results not presented). There is a perception among many sheep breeders in the UK that the wild type ARQ allele, which is known to be associated with high to moderate susceptibility to scrapie, is also associated with improved fitness under harsh environmental conditions (Nicholls et al., 2006). Our results did show that ARQ carrying ewes performed differently from others, with ARQ carrier ewes at least 0.5 kg lighter than ARQ noncarriers, and ARQ heterozygotes tending to mobilize more fat reserves than noncarriers. Efficient tissue mobilization is necessary for meeting nutrient demand at times of deficiency such as during lactation and may affect performance and fitness of the reared offspring (Russel et al., 1968; Lambe et al., 2007). In dairy cattle, higher producing cows have also been shown to lose more body reserves than lower producing ones during periods of negative energy balance (Coffey et al., 2004).
In summary, breeding programs aiming to reduce scrapie through the PrP gene are being implemented in many countries. Selecting only against the VRQ allele would have no effect on any of the ewe traits studied. However, more stringent programs selecting against the ARQ allele may also affect the overall BW of the ewes and their ability to mobilize body reserves at times of nutrient deficiency. Selecting for non-ARQ carrier ewes might result in heavier ewes with less ability to mobilize body reserves of nutrients when needed.
|
Footnotes
|
|---|
1 This work was funded by Defra (Department for Environment, Food and Rural Affairs, UK). The authors are grateful to W. Hill (University of Edinburgh), S. Bishop (RI), G. Simm (SAC), and K. Boulton (MLC) for reviewing and commenting on this manuscript, to J. Connington, A. McLaren, K. McLean, and M. Steel (SAC) for providing the data and for MLC (UK), ST (UK), and RBST (UK) for in-kind contribution. 
2 Corresponding author: Rami.Sawalha{at}sac.ac.uk
Received for publication October 11, 2007.
Accepted for publication April 7, 2008.
|
LITERATURE CITED
|
|---|
Álvarez, L., B. Gutiérrez-Gil, F. San Primitivo, L. F. de la Fuente, and J. J. Arranz. 2006. Influence of prion protein genotypes on milk production traits in Spanish Churra sheep. J. Dairy Sci. 89:1784–1791.[Abstract/Free Full Text]
Casellas, J., G. Caja, R. Bach, O. Francino, and J. Piedrafita. 2007. Association analyses between the prion protein locus and re-productive and lamb weight traits in Ripollesa sheep. J. Anim. Sci. 85:592–597.[Abstract/Free Full Text]
Coffey, M. P., G. Simm, J. D. Oldham, W. G. Hill, and S. Brotherstone. 2004. Genotype and diet effects on energy balance in the first three lactations of dairy cows. J. Dairy Sci. 87:4318–4326.[Abstract/Free Full Text]
Conington, J., S. C. Bishop, B. Grundy, A. Waterhouse, and G. Simm. 2001. Multi-trait selection indexes for sustainable UK hill sheep production. Anim. Sci. 73:413–423.
Conington, J., S. C. Bishop, N. R. Lambe, L. Bunger, and G. Simm. 2006. Testing selection indices for sustainable hill sheep production - lamb growth and carcass traits. Anim. Sci. 82:445–453.[CrossRef]
de Vries, F., N. Borchers, H. Hamann, C. Drögemüller, S. Reinecke, W. Lüpping, and O. Distl. 2004. Associations between the pri-on protein genotype and performance traits of meat breeds of sheep. Vet. Rec. 155:140–143.[Abstract/Free Full Text]
de Vries, F., H. Hamann, C. Drögemüller, and O. Distl. 2006. Associations between prion protein genotype and type traits in East Friesian milk sheep. Vet. Rec. 158:849–852.[Abstract/Free Full Text]
de Vries, F., H. Hamann, C. Drögemüller, M. Ganter, and O. Distl. 2005. Analysis of associations between the prion protein genotypes and production traits in East Friesian milk sheep. J. Dairy Sci. 88:392–398.[Abstract/Free Full Text]
Eglin, R. D., R. Warner, S. Gubbins, S. K. Sivam, and M. Dawson. 2005. Frequencies of PrP genotypes in 38 breeds of sheep sampled in the National Scrapie Plan for Great Britain. Vet. Rec. 156:433–437.[Abstract/Free Full Text]
Gilmour, A., B. Cullis, S. Welham, B. Gogel, and R. Thompson. 2004. An efficient computing strategy for prediction in mixed linear models. Comput. Stat. Data Anal. 44:571–586.[CrossRef]
Gilmour, A. R., B. J. Gogel, B. R. Cullis, S. J. Welham, and R. Thompson. 2002. ASReml User Guide Release 1.0. VSN Int. Ltd., Hemel Hempstead, UK.
Goldmann, W., N. Hunter, G. Smith, J. Foster, and J. Hope. 1994. PrP genotype and agent effects in scrapie - change in allelic interaction with different isolates of agent in sheep, a natural host of scrapie. J. Gen. Virol. 75:989–995.[Abstract/Free Full Text]
Isler, B. J., B. A. Freking, R. M. Thallman, M. P. Heaton, and K. A. Leymaster. 2006. Evaluation of associations between prion haplotypes and growth, carcass, and meat quality traits in a Dorset x Romanov sheep population. J. Anim. Sci. 84:783–788.[Abstract/Free Full Text]
Jamrozik, J., L. R. Schaeffer, and J. C. M. Dekkers. 1997. Genetic evaluation of dairy cattle using test day yields and random regression model. J. Dairy Sci. 80:1217–1226.[Abstract]
Lambe, N. R., S. Brotherstone, M. J. Young, J. Conington, and G. Simm. 2005. Genetic relationships between seasonal tissue levels in Scottish Blackface ewes and lamb growth traits. Anim. Sci. 81:11–22.
Lambe, N. R., J. Conington, K. A. McLean, L. Bunger, and G. Simm. 2007. Relationships between mobilisation of body reserves in hill ewes and lamb production to weaning. Page 118 in Proc. Brit. Soc. Anim. Sci., Southport, UK.
Lambe, N. R., G. Simm, M. J. Young, J. Conington, and S. Brotherstone. 2004. Seasonal changes in tissue weights in Scottish Blackface ewes over multiple production cycles. Anim. Sci. 79:373–385.
Lambe, N. R., M. J. Young, S. Brotherstone, T. Kvame, J. Conington, K. Kolstad, and G. Simm. 2003a. Body composition changes in Scottish Blackface ewes during one annual production cycle. Anim. Sci. 76:211–219.
Lambe, N. R., M. J. Young, K. A. McLean, J. Conington, and G. Simm. 2003b. Prediction of total body tissue weights in Scottish Blackface ewes using computed tomography scanning. Anim. Sci. 76:191–197.
Lewis, R. M., and S. Brotherstone. 2002. A genetic evaluation of growth in sheep using random regression techniques. Anim. Sci. 74:63–70.
Meyer, K. 2000. Random regressions to model phenotypic variation in monthly weights of Australian beef cows. Livest. Prod. Sci. 65:19–38.[CrossRef][Medline]
Meyer, K., and W. G. Hill. 1997. Estimation of genetic and phenotypic covariance functions for longitudinal or ‘repeated’ records by restricted maximum likelihood. Livest. Prod. Sci. 47:185–200.[CrossRef]
Nicholls, N., L. Kruuk, M. Woolhouse, E. Stevenson, M. Gravenor, and M. Baylis. 2006. Investigation of farmer regard for scrapie susceptible sheep. Vet. Rec. 158:732–734.[Free Full Text]
Ponz, R., M. T. Tejedor, L. V. Monteagudo, and M. V. Arruga. 2006. Scrapie resistance alleles are not associated with lower prolificity in Rasa Aragonesa sheep. Res. Vet. Sci. 81:37–39.[CrossRef][Medline]
Russel, A. J. F., J. M. Doney, and R. G. Gunn. 1969. Subjective assessment of body fat in live sheep. J. Agric. Sci. 72:451–454.
Russel, A. J. F., R. G. Gunn, and J. M. Doney. 1968. Components of weight loss in pregnant hill ewes during winter. Anim. Prod. 10:41–53.[Medline]
SAC. 2007. The farm management handbook. The Scottish Agricultural College, Edinburgh, UK.
Sawalha, R. M., S. Brotherstone, J. Conington, and B. Villanueva. 2007a. Lambs with scrapie susceptible genotypes have higher postnatal survival. PLoS One 2:e1236. doi:10.1371/journal.pone.0001236[CrossRef]
Sawalha, R. M., S. Brotherstone, W. Y. N. Man, J. Conington, L. Bünger, G. Simm, and B. Villanueva. 2007b. Associations of polymorphisms of the ovine prion protein gene with growth, car-cass, and computerized tomography traits in Scottish Blackface lambs. J. Anim. Sci. 85:632–640.[Abstract/Free Full Text]
Shaffer, J. P. 1995. Multiple hypothesis testing. Annu. Rev. Psychol. 46:561–584.[CrossRef]
Simm, G., R. M. Lewis, J. E. Collins, and G. J. Nieuwhof. 2001. Use of sire referencing schemes to select for improved carcass composition in sheep. J. Anim. Sci. 79(E. Suppl.):E255–E259.[Abstract/Free Full Text]
Vitezica, Z. G., C. R. Moreno, L. Bodin, D. François, F. Barillet, J. C. Brunel, and J. M. Elsen. 2006. No associations between PrP genotypes and reproduction traits in INRA 401 sheep. J. Anim. Sci. 84:1317–1322.[Abstract/Free Full Text]
Vitezica, Z. G., C. R. Moreno, J. Bouix, F. Barillet, G. Perret, and J. M. Elsen. 2005. A study on associations between PrP genotypes and meat traits in French sheep breeds. Anim. Sci. 81:325–330.[CrossRef]
Weber, M. L., and J. M. Thompson. 1998. Seasonal patterns in food intake, live mass, and body composition of mature female fallow deer (Dama dama). Can. J. Zool. 76:1141–1152.[CrossRef]
Top
Abstract
INTRODUCTION
), and standard deviations multiplied by 5 (
) of computer tomography (CT) tissue weights, body condition score, and BW by recoding event and age. Animals were recorded 4 times per year (M = mating, P = pregnancy, L = lactation, and W = weaning) for 4 yr (from 2 to 5 yr of age).