Gene polymorphisms associated with temperament in Merino sheep

When individuals are exposed to stressful environmental challenges, the response varies widely in one or more of three components of the stress response: psychological (emotional), behavioural and physiological. This variability among individuals can be defined as temperament. Temperament affects not only the perception and responses to the stressor, but also can affect cognition. The genetic basis of temperament is not clearly understood, but it could be explained by polymorphisms in genes that regulate brain activity and glucocorticoid synthesis. In this thesis, I tested the general hypothesis that polymorphisms of specific genes will be associated with the difference in behavioural reactivity and physiological response in sheep of known temperament (nervous or calm). I mainly used sheep from the UWA Merino flock that had been selected over 20 generations for two temperaments: a nervous line and a calm line, based on high or low reactivity to humans and social isolation.
Firstly, using nervous (n = 58) or calm (n = 59) ewes from lines that had been selected for temperament for 20 generations, we identified polymorphisms in a specific gene responsible for cortisol production (CYP17), and in three specific genes associated with personality and behavioural traits: dopamine receptors 2 and 4 (DRD2, DRD4), and monoamine oxidase A (MAOA). The frequencies of CYP17 SNP628 and DRD2 SNP939 differed significantly between the two temperament lines, but those for DRD2 SNP483 and the 2 MAOA SNP genotypes did not.
To validate the correctness of the two SNPs associated with temperament (CYP17 SNP628 and DRD2 SNP939) as genetic markers for prediction of sheep temperament, we genotyped 278 non-selected sheep for the DRD2 SNP939 and CYP17 SNP628 polymorphisms. We then tested their behavioural response to the challenges used to assess temperament) and their cortisol response to exogenous ACTH. The results showed that responses to the behavioural tests (‘arena test’ and ‘isolation box’ test) were affected by the DRD2 SNP939 genotype and the cortisol response to ACTH challenge was affected by the CYP17 SNP628 genotype. We concluded that, for sheep, a combination of the DRD2 SNP939 C allele and the CYP17 SNP628 A/A genotype could be a genetic marker for prediction of nervous temperament, and that a combination of DRD2 SNP939 T/T and CYP17 SNP628 G/G could be a genetic marker for prediction of calm temperament.
We then calculated the heritability of behavioural reactivity to isolation (one measurement of temperament) based on the 2 SNPS genotypes associated with temperament phenotype. The outcome suggested that the phenotypic heritability of reactivity to social isolation, as measured by the agitation score, can be improved by genetic selection based on these 2 SNPs genotypes, and that the combination of the 2 SNPs genotypes was associated with the highest heritability. This result confirmed the above conclusions that DRD2 SNP939 and CYP17 SNP628 genotypes are associated with the two temperaments.
In addition to the association of emotional reactivity (or ‘temperament’) with behavioural and physiological responses, previous studies have suggested that temperament is associated with cognition. Moreover, polymorphism in DRD2 has been implicated in variability in cognitive functions in humans. However, the effect of the interaction between emotional reactivity and DRD2 polymorphism on cognitive function is poorly understood. Therefore, we investigated the association of our temperament-associated DRD2 SNP939 genotypes with cognitive learning abilities in sheep and found that the ‘calm’ temperament-associated DRD2 SNP939 T/T genotype was linked to a greater ability in reversal learning of spatial discrimination than the ‘nervous’ temperament-associated C/C +C/T genotype.
In conclusion, this thesis is a first step in the identification of potential genetic markers for prediction of calm or nervous temperament in sheep. The DRD2 SNP939 is specifically associated with behavioural reactivity and cognitive learning ability, and the CYP17 SNP628 is specifically associated with the physiological cortisol response. The combination of these 2 SNPs could be used as a genetic marker for the temperament prediction. However, many other genes involved in brain neurotransmitter systems and in hypothalamic-pituitary-adrenal axis could also be associated with sheep temperament, so, perhaps using genome-wide association study (GWAS), we need to investigate the association of other genes with sheep temperament.