Adaptation to a constantly changing environment is fundamental to every living organism. The hypothalamic–pituitary–adrenocortical (HPA) axis is a key component of the adaptation process. The present study tests the hypothesis that vasopressin (AVP) is required for the HPA response to acute stimuli. To accomplish this, naturally AVP-deficient Brattleboro rats were exposed to a wide range of stimuli and their HPA response was compared with heterozygous littermattes. The circadian rhythmicity of plasma ACTH and corticosterone was not different between the two genotypes. The ACTH and corticosterone response to volume load, restraint or aggressive attack were decreased in AVP-deficient rats. The stress-induced increase in ACTH, but not corticosterone, was significantly impaired in AVP-deficient animals after novelty, elevated plus-maze, forced swim, hypoglycaemia, ulcerogenic cold immobilisation, lipopolysaccharide, hypertonic saline and egg white injection. The HPA response to social avoidance, ether inhalation and footshock was not different between the genotypes. In vitro, the hypophysis of AVP-deficient animals showed a reduction in stimulated ACTH production and their adrenal glands were hyporeactive to ACTH. A dissociation between the ACTH and corticosterone response was observed in several experiments and could not be explained by an earlier ACTH peak or enhanced adrenal sensitivity, suggesting the existence of paraadenohypophyseal neuroendocrine regulators. Loss of AVP affected the HPA response to a wide variety of stressors. Interestingly, the contribution of AVP to the HPA response was not specific for, nor limited to, a known stressor category. Thus, there is a context-specific requirement for AVP in stress-induced activation of the HPA axis.

Stress adaptation is of utmost importance for the maintenance of homeostasis and, therefore, of life itself. The prevalence of stress-related disorders is increasing, emphasizing the importance of exploratory research on stress adaptation. Two major regulatory pathways exist: the hypothalamic–pituitary–adrenocortical axis and the sympathetic adrenomedullary axis. They act in unison, ensured by the enormous bidirectional connection between their centers, the paraventricular nucleus of the hypothalamus (PVN), and the brainstem monoaminergic cell groups, respectively. PVN and especially their corticotropin-releasing hormone (CRH) producing neurons are considered to be the centrum of stress regulation. However, the brainstem seems to be equally important. Therefore, we aimed to summarize the present knowledge on the role of classical neurotransmitters of the brainstem (GABA, glutamate as well as serotonin, noradrenaline, adrenaline, and dopamine) in stress adaptation. Neuropeptides, including CRH, might be co-localized in the brainstem nuclei. Here we focused on CRH as its role in stress regulation is well-known and widely accepted and other CRH neurons scattered along the brain may also complement the function of the PVN. Although CRH-positive cells are present on some parts of the brainstem, sometimes even in comparable amounts as in the PVN, not much is known about their contribution to stress adaptation. Based on the role of the Barrington’s nucleus in micturition and the inferior olivary complex in the regulation of fine motoric—as the main CRH-containing brainstem areas—we might assume that these areas regulate stress-induced urination and locomotion, respectively. Further studies are necessary for the field.

In adulthood the hypothalamo-pituitary-adrenal axis is controlled by both CRH and arginine vasopressin (AVP). However, in neonates CRH secretion is very low, whereas AVP secretion is fully functional. This suggests that the role of AVP is more pronounced in young than in adult rats. We investigated the role of AVP by studying stress responses in 5, 10, and 20-d-old AVP-deficient Brattleboro rats. Two different stressors were applied: 24-h maternal separation and Hypnorm Grove Oxford UK injections. In heterozygous controls (that do express AVP), both stressors increased plasma ACTH and corticosterone. The ACTH stress response disappeared in AVP-deficient rats, demonstrating that during the perinatal period, the secretion of this hormone is controlled by AVP. Surprisingly, corticosterone responses remained intact in AVP-deficient rats. Similar findings were obtained after 1-, 4-, 12-, and 24-h long maternal separations. Thus, preserved corticosterone stress responses were not explained by changes in the timing of ACTH secretion. In vitro experiments suggested that the dissociation of ACTH and corticosterone stress responses can only be partly explained by higher ACTH responsiveness of the adrenal cortex in AVP-deficient rats. Together, our results show that in neonatal periods, AVP is crucial for the expression of ACTH stress responses, but neither AVP nor ACTH is necessary for the induction of corticosterone stress responses. Discrepant ACTH and corticosterone stress responses may reflect compensatory mechanisms activated by AVP deficiency, but disparate findings suggest that they rather depict a neonate-specific mechanism of hypothalamo-pituitary-adrenal-axis control.

Diabetes mellitus (DM), as chronic stress activates the hypothalamo-pituitary-adrenocortical axis. We examined whether arginine vasopressin (AVP) and the hypothalamic paraventricular nucleus (PVN) participate in DM-induced chronic stress symptoms. AVP-deficient Brattleboro or PVN-lesioned Wistar rats were used with heterozygous or sham-operated controls. The rats were studied 2 wk after a single injection of streptozotocin. The appearance of DM (enhanced water consumption and blood glucose elevation) and the chronic stress-like somatic changes (body weight decrease, thymus involution, adrenal gland hypertrophy) were not influenced by the lack of AVP. By contrast, PVN lesion significantly attenuated DM-induced thymus involution and adrenal gland hypertrophy as well as the increase in water consumption. The corticotropin-releasing hormone mRNA in PVN was diminished by DM and elevated by the lack of AVP without interaction. DM elevated the proopiomelanocortin (POMC) mRNA in the anterior lobe of the pituitary. The lack of AVP had no effect, whereas lesioning the PVN significantly diminished the elevation. The elevated basal corticosterone plasma levels detectable in DM were influenced neither by the lack of AVP nor by lesioning the PVN. Thus the lack of AVP had no influence on DM-induced chronic stress symptoms, but lesioning the PVN attenuated part of them. However, the lack of elevation in POMC mRNA after PVN lesion, together with the maintained corticosterone elevation, suggests that direct adrenal gland activation occurs in untreated DM.