One of the main hypotheses about the pathogenesis of depression concerns hyperactivity of the hypothalamo-pituitary-adrenal (HPA) axis. This axis is driven by corticotropin-releasing hormone release of neurons located in the paraventricular nucleus (PVN) of the hypothalamus, that causes adrenocorticotropic hormone (ACTH) release at the level of the pituitary. ACTH release is potentiated by hypothalamic vasopressin (AVP).1 AVP differs only two amino acids from oxytocin (OXT), a neuropeptide with many effects in social interactions.2 Both neuropeptides are released from the hypothalamic paraventricular and supraoptic nucleus (SON).1 Whereas AVP potentiates HPA-axis activity,3 animal experiments have shown that OXT attenuates the stress-induced activity of the HPA-axis in various species, including humans,4 and that OXT inhibits basal HPA-axis activity.5 Van Londen et al.6 found elevated AVP plasma levels in depressed patients and normal OXT levels, but described a larger variability in these levels compared to controls. Plasma OXT does not readily cross the blood–brain barrier, and there is no direct relationship between the release of OXT into the blood by the neuropituitary and the variations in OXT levels in the cerebrospinal fluid.1, 2 We therefore previously determined the number of OXT-expressing neurons in the PVN of depressed patients and this number turned out to be increased.7
Recently, we found in a post-mortem sample of depressed subjects a significant increase of AVP mRNA expression in the SON, and in both the SON and PVN when only the more severe, melancholic subgroup was taken into account.8 In the same group of depressed patients, with one control added, we performed a quantitative OXT mRNA in situ hybridization using the same technique.8, 9 Briefly, hypothalami of depressed subjects (six melancholic type, three non-melancholic type) and nine control subjects matched for age and sex were obtained from the Netherlands Brain Bank in accordance with the formal protocols for use of human brain material and clinical information for research purposes.8 Differences among groups were evaluated by the non-parametric Kruskal–Wallis test and Mann–Whitney U-test. Correlations were evaluated by Spearman's rho. Statistical significance was set at P<0.05.
A significant increase of OXT mRNA in melancholic type patients compared to non-melancholic type patients existed in the PVN (Z=−2.074, P=0.038), whereas melancholic type patients compared to controls showed a trend (P=0.099) towards higher OXT mRNA in the PVN (Figure 1). There was no difference in OXT mRNA in either the PVN or SON when comparing the entire group of depressed patients with control subjects. The group of depressed patients did not differ significantly from the control subjects concerning gender, post-mortem delay and fixation time.
This is the first report in which OXT mRNA has been quantified in the SON and PVN in depressed patients and control subjects. In the PVN, we found an increased OXT mRNA expression in the melancholic patients compared to the non-melancholic subgroup. Recently, in the same subgroup of patients we found an increase in AVP mRNA when comparing melancholic to non-melancholic subtype of depression (P=0.02, data not shown), and also the entire depressed group showed a significant increase in AVP mRNA expression in the SON compared to controls.8 Furthermore, a correlation existed between the amount of AVP mRNA8 and the amount of OXT mRNA in the PVN (rho=0.870, P=0.002) and SON (rho=0.745, P=0.021) in depressed subjects, whereas in controls such a correlation was found in the SON (rho=0.786, P=0.021).
When interpreting the results from the present experiment we have to be cautious because of the small groups. The increase in OXT mRNA in melancholically depressed patients could be related to specific symptoms of melancholic type of depression, like weight loss and loss of appetite, as OXT is a satiety peptide.1 Supportive of this hypothesis, we found the highest amount of OXT mRNA in a melancholically depressed subject, whose medical record mentioned 16 kg weight loss during a depressive episode during the last year of her life.
In conclusion, our data, although from a small group, indicate that a distinction should be made between patients with and without the melancholic subtype of depression, when evaluating molecular changes in the brain.