Hormonal interactions within the hypothalamus and pituitary with respect to stress and reproduction in sheep.

Endocrine systems may be used as indicators of stress in two ways. The primary role of a hormone may be as part of the homeostatic response to a stimulus (e.g., adrenaline, corticosteroids). The amplitude of hormone response may correlate with the severity of the stimulus and any change indicate that the body is responding. Alternatively, a hormone may have a key role in normal body function (e.g., reproduction) and stress may deleteriously alter the hormone signal prevent normal function. This demonstrates that the stimulus was sufficiently severe that homeostatic mechanisms were unable to maintain normal function. Stress may effect reproduction by reducing both LH pulse amplitude and frequency. The LH surge may also be delayed. Several mechanisms may account for these effects both at the hypothalamus and pituitary. Corticosteroids have a broad, yet fundamental, role in homeostasis and have been used as primary indicators of stress for many years. Excess corticosteroid can be detrimental so the concentration is controlled via the hypothalamus-pituitary-adrenal (HPA) axis by multi-level feedback mechanisms. Under field and experimental conditions, after an initial large response prolonged stimulation leads to a gradually reducing plasma corticosteroid concentrations. This has been interpreted as a reduction in perceived stimulus severity or habituation to the stimulus and the animal deemed "less stressed" and its welfare "better." However, this reduction may be due to the intrinsic control mechanisms designed to prevent prolonged increases in corticosteroid concentrations. The stress signal at higher brain levels may still be present and the animal may still be experiencing the stimulus as aversive. Thus, the welfare interpretation of a corticosteroid concentration may differ during the time course of a stress response. A greater understanding of the mechanisms controlling corticosteroid secretion at each level of the HPA is required to determine what is the correct interpretation at any time point. To address these issues, we have used mathematical modelling to produce representations of possible control mechanisms at each level of the HPA. The starting point was to measure AVP and CRH concentrations in hypophysial portal blood and ACTH and cortisol concentrations in jugular blood in conscious sheep during 2h road transport (a cognitive stimulus). Modelling identified the signal inputs that were most likely to explain the secretion rate of each hormone. Modelling suggested that the reduction in AVP and CRH secretion observed during transport was most likely due to a reduction in stimulus input, with a significant contribution from cortisol negative feedback only on AVP secretion. At the pituitary level, ACTH secretion was stimulated more by AVP than by CRH (ratio 2.3:1) and there was also a stimulatory effect related to cortisol concentration at the time of sampling. However, the responses to both stimuli were curtailed by cortisol negative feedback and an inhibitory effect of prior CRH concentration. These are complex effects, but the modelling does suggest that while "stress" inputs may reduce over time hormone negative feedback is a major factor reducing hormone responses. When interpreting hormone data for animal welfare purposes, it is important not to interpret a reduction in hormone concentration due to intrinsic hormone control mechanisms as a reduction due to a decrease in the stress stimulus.