Central and peripheral aspects of hypothalamic-pituitary-adrenal (HPA) axis dysfunction

An aberrant regulation of the hypothalamic-pituitary-adrenal (HPA)
axis is closely associated with the pathophysiology of affective
disorders such as major depression (MD). Accordingly, patients
suffering from MD frequently show profound neuroendocrine
alterations with hyper- or hypo-cortisolism as a result of a
dysregulated stress hormone system. Focussing on this key
endophenotype of MD, the ‘stress reactivity’ (SR) mouse model was
recently established, consisting of three independent mouse lines,
the high (HR), intermediate (IR) and low (LR) stress reactivity
line, selectively bred for differences in their corticosterone
(CORT) secretion in response to a psychological stressor. Previous
studies revealed distinct differences between HR, IR and LR animals
regarding sleep architecture, activity rhythms, emotional
behaviour, cognition as well as neuroendocrine functions,
resembling several endophenotypes observed in depressed patients.
In the series of studies presented in this work, we aimed to
investigate whether the differences between HR, IR and LR mice were
restricted to the peripheral phenomenon of adrenal CORT secretion,
or whether these endophenotypes were brought about by an aberrant
regulation of upstream control centres of the HPA axis. To this
end, we performed experiments investigating all functional levels
of HPA axis control, i.e. the adrenals, the pituitary and brain
centres known to be involved in the neuroendocrine stress response.
Moreover, we assessed the expression of corticosteroid-binding
globulin (CBG), which contributes to the transport and delivery of
CORT to its target tissues. Finally, we studied HPA axis regulatory
mechanisms by means of the combined
dexamethasone/corticotropin-releasing hormone (Dex/CRH) test. At
the level of the adrenal, we found that pharmacological inhibition
of the biosynthesis and secretion of CORT using metyrapone had a
significant impact on the stress-coping behaviour of HR, IR and LR
animals as determined in the forced swim test (FST). As another
peripheral factor influencing the secretion of CORT, we assessed
the adrenal sensitivity of the animals to adrenocorticotropic
hormone (ACTH) in vivo. After a Dex-mediated inhibition of
endogenous ACTH release from the anterior pituitary, LR animals
showed a markedly reduced CORT surge compared to HR mice in
response to a stimulation of the adrenals with two doses of
exogenous ACTH, indicating an enhanced adrenal sensitivity in HR
mice and a blunted responsiveness to ACTH in LR mice. In addition,
we found significant differences in plasma CBG levels between the
three mouse lines (HR>IR>LR), concomitant with differences in
free plasma CORT both, basal and in response to 15 min restraint
stress (HR>IR>LR). Since only free CORT is biologically
active, these results indicate that CBG might play a role in the
endophenotypes of the SR mouse lines. At the pituitary level, we
detected significantly altered ACTH protein levels (HR>IR≥LR)
and proopiomelanocortin mRNA expression (HR>IR>LR),
suggesting a differential activation of the anterior pituitary
between the three lines, which is in line with the observed
differences in stress reactivity. In the brain, we assessed the
neuronal activation induced by an acute stressor in regions known
to be involved in HPA axis function such as the prefrontal cortex,
the basolateral amygdala, the hippocampus and the paraventricular
nucleus of the hypothalamus (PVN) using c-fos in-situ
hybridisation. Acute stress exposure markedly increased neuronal
activation in all investigated brain areas. However, significant
differences in the neuronal excitation between the three lines were
only detected in the PVN (HR>IR>LR), indicating an altered
activation of the animals’ HPA system orchestrated by this nucleus.
Finally, we assessed HPA axis regulatory mechanisms by means of the
Dex/CRH test. Our results revealed considerable similarities to
clinical studies, with HR mice showing signs of Dex non-suppression
in addition to an overshooting CORT surge after CRH stimulation,
mirroring the HPA axis hyper-active state of patients suffering
from the psychotic or melancholic subtype of MD. In contrast, LR
animals presented a strong Dex-induced CORT suppression and a
blunted response to the CRH stimulation, resembling the situation
observed in atypically depressed patients. Strikingly, chronic
fluoxetine (Flx) treatment enhanced the negative feedback
regulation of the HPA axis in all three lines of the SR mouse
model. In particular, the Dex-mediated CORT suppression in HR mice
was restored by Flx. Furthermore, Flx affected the stress-coping
behaviour of the animals in the FST. Flx induced a reduction in
active coping, indicating an attenuation of the hyper-aroused
state, particularly in HR mice. Taken together, the series of
studies presented here demonstrated that the SR mouse model shows
functional alterations on all levels of the HPA axis - peripheral,
central and regarding the regulation – similar to the
endophenotypes of MD patients, thus revealing a high level of face
and construct validity of the model. Hence, the SR mouse model can
serve as a valuable tool in the discovery and validation of new
drug targets and improve already existing treatments of MD,
particularly those targeting the HPA system.