Prediction of prion protein genotype and association of this genotype with lamb performance traits of Suffolk sheep1,2

Abstract

The association of the prion protein (PrP) gene with susceptibility to scrapie has formed the basis of selection programs aimed at eradicating the disease from sheep populations. Animals are genotyped for the PrP gene and those with the less susceptible genotypes are selected. The objectives of this study were to determine the effectiveness of predicting PrP genotypes by using information from relatives and to investigate the association of the PrP genotype with lamb performance traits in Suffolk sheep. Data were obtained from a scrapie-affected flock maintained in Scotland. A total of 643 were animals genotyped at codon 171 of the PrP gene with 2 alleles, R and Q. The genotypes of these animals were used to predict the genotypes of 5,173 nongenotyped animals in the same flock using segregation analysis. The genotype of nongenotyped animals was predicted from the probabilities for each possible genotype; further, an overall index for each animal was calculated to reflect the accuracy of prediction. Association analyses of the PrP gene (using animals with both known and inferred genotypes) with BW at birth, at weaning (56 d), and at 150 d, and for backfat and muscle depths at 150 d of age were carried out. A linear mixed model with random direct and maternal additive genetic effects, maternal permanent and temporary environmental effects, and year of birth was tested, and the most appropriate model was used for each trait. The expected number of Q alleles carried (from 0 to 2) by each animal was calculated and used in the model as a linear and quadratic covariate to test for associations with possible additive and dominance PrP gene effects, respectively. Results showed that the genotypes of relatively few animals (235) were inferred with certainty (compared with the 5,173 nongenotyped animals). Approximately 25% of the 5,173 predicted genotypes were inferred with a genotype probability index of 50% and greater. There was no significant association of the PrP gene with any of the performance traits studied (there were no significant additive or dominance effects). Such was the case whether data on animals with known or with both known and predicted genotypes were considered. It can be concluded that selection for PrP-resistant alleles in Suffolk sheep is unlikely to affect performance directly.