Behavioural and physiological responses to stressors in sheep with temperament classified by genotype or phenotype

Abstract

The single nucleotide polymorphism (SNP) rs107856856, located in the tryptophan hydroxylase-2 gene, is associated with the behavioural phenotype for sheep temperament measured at weaning. Here, we tested the association between that SNP and physiological and behavioural responses to stressors in adult sheep. Two groups of adult sheep, one with genotype A/A (calm genotype) and the other with G/G (nervous genotype) in rs107856856, were selected from 160 sheep and were exposed, twice, to an open-field arena and an isolation box test (IBT). During each repeat, the behaviour and physiological responses (cortisol, prolactin, dehydroepiandrosterone [DHEA], brain derived neurotrophic factor [BDNF], characteristics of the response of body temperature, and oxidative stress) were measured. The behavioural and physiological responses of the sheep were compared between genotypes and also between groups classified on their phenotype as assessed by their initial isolation box score ("low responders" and "high responders"). The SNP rs107856856 had some effects on the behavioural phenotype (IBT score) but no effects on the physiological response to stress (cortisol, prolactin, DHEA, BDNF, oxidative stress or changes in body temperature) in the adult sheep, probably because the sheep were exposed, and therefore had adapted, to human contact during their life.

Figure 1

Scores for number of bleats, and number of crosses during the arena test, IBT, and IBT-LWT (standardised to the body weight of each animal) in sheep with the Calm A/A (white bar) or the Nervous G/G (black bar) genotype for SNP rs107856856, and in sheep that were phenotypically classified as low responder (grey bar) or high responder (dark grey bar) in two sessions (S1 = 1st session, S2 = 2nd session) of the behavioural tests. Data are shown as mean ± SE. * indicates a significant difference P < 0.05.

Figure 2

Plasma concentrations, at the end of each session (S1 = 1st session, S2 = 2nd session) of the behavioural tests, of cortisol (upper strip), prolactin (second strip), DHEA (third strip), and BDNF (lower strip) in sheep carrying the Calm A/A (white bar) or the Nervous G/G (black bar) genotype for SNP rs107856856, and in sheep that were phenotypically classified as low responder (grey bar) or of high responder (dark grey bar) in two sessions (S1 = 1st session, S2 = 2nd sesision) of the behavioural tests. Data are shown as mean ± SE. * indicates a statistical difference P < 0.05.

Figure 3

Level of blood oxidative stress of all the experimental sheep (there was no effect of either genotype or phenotype) before (white bar) and after (black bar) each of the behavioural tests during the two sessions (S1 = 1st session; S2 = 2nd session). Oxidative stress was calculated as the ratio of oxidised thiol groups on albumin to the non-oxidised groups (mean ± SE). Means with different superscripts are statistically different at P < 0.05.

Figure 4

Area under the curve (AUC; top strip), and amplitude (lower strip) of the stress induced hyperthermia (mean ± SE) in response to behavioural testing during the two sessions (S1 = 1st session; S2 = 2nd session),and in response to the presence or absence of dogs (absence = absence of dogs; presence = presence of dogs) in sheep carrying the Calm A/A (white bar) or the Nervous G/G (black bar) genotypes for SNP rs107856856, and in sheep that were phenotypically classified as either low responder (grey bar) or high responder (dark grey bar) Bars show mean ± SE. Means with different superscripts within a graph indicate statistical difference at P < 0.05.

Figure 5

Correlation between each of the behavioural and physiological responses to stress that were measured in each of the two sessions of behavioural tests of all animals regardless of groups (genotype or phenotype) except for animals that had a missing variable. AUC the area under the curve of the stress induced hyperthermia (SIH), IBT isolation box test, OXYPre the blood oxidative stress before each behavioural test, OXYPost the blood oxidative stress after each behavioural test, DHEA dehydroepiandrosterone, BDNF brain derived neurotrophic factor, S1 the first session of behavioural tests, S2 the second session of behavioural tests. The size of circle and the colour scale indicate the coefficient values of each combination (as given in the legend below the figure). * indicates a significant correlation at P < 0.05.

Figure 6

The activity score in sheep carrying the Calm A/A (white bar) or the Nervous G/G genotypes (black bar) for SNP rs107856856, and in sheep that were phenotypically classified as low responder phenotype (grey bar) or of high responder phenotype (dark grey bar) in two sessions (S1 = 1st session; S2 = 2nd session), and in response to the presence or absence of dogs (absence = absence of dogs; presence = presence of dogs). Data are shown as mean ± SE. There was no effect of either genotype pr phenotype, and so means with different superscripts within a graph indicate a statistical difference between periods (S1, S2, absence, and presence) at P < 0.05.

Figure 7

Characteristics of the circadian rhythm of body temperature in sheep carrying the Calm A/A (white bar) or the Nervous G/G (black bar) genotypes for SNP rs107856856, and in sheep that were phenotypically classified as either low responder (grey bar) or high responder (dark grey bar) (mean ± SE, n = 7). From left to right, the panels show amplitude, mesor, cosinor minimum, cosinor maximum over three days on five occasions from prior to the first behavioral test day (Period 1; P1), the three days of immediately after the first behavioral test day (Period 2; P2), the three days immediately after that (Period 3; P3), the three days immediately after the second behavioral test day (Period 4 ; P4), and finally the 3 days after that (Period 5 ; P5), as illustrated on Fig. 8. Bars show mean ± SE. * indicates P < 0.001 in the pairwise t test to P1.

Figure 8

Experimental timeline and representative profile of core body temperature in one sheep for the duration of the experiment. The plot shows the original record of 5-min recordings of T_c (black line), and fitted cosinor for each 24 h (red line). The black numbered boxes above the plot indicate the days when behavioural phenotype was assessed. The stress induced hyperthermia (SIH) that was associated with each stressor is indicated by arrows (behavioral test) and triangle (exposure to dog). Each period of three days when the characteristics of the circadian rhythm of core body temperature (CRT) was analysed is indicated by a numbered open box above the trace (P1 etc.)