Genetic factors are known to influence individual vulnerability to stress and thus to stress-related diseases including affective disorders, metabolic troubles, addiction to drugs, aging or auto-immune diseases. So far the genes identified as influencing stress responses are few and they stem from candidate gene approaches, e.g. the genes encoding for the glucocorticoid receptor, the serotonin transporter or the neuropeptide Y, all known to be primary mediators of the stress response. It is from a non-hypothesis driven strategy that we have originally identified the gene encoding transcortin as a genetic factor influencing the glucocorticoid response to stress. Transcortin also known as corticosteroid binding globulin (CBG) is a glycoprotein binding to glucocorticoids (cortisol in human, corticosterone in rodents) with a high affinity in the blood. Transcortin is well-known as the glucocorticoid carrier, playing a role is glucocorticoid bioavailability as only the unbound free glucocorticoid hormone is active on target cells. In the article presented here, we “loop the loop” by showing that changing specifically transcortin gene expression alters stress responses in mice.

We have produced mice deleted totally or partially for the gene Cbg encoding for transcortin. Cbg-/- mice have no transcortin at all whereas Cbg+/- display 50% of transcortin gene expression levels compared to the control animals Cbg+/+. We first looked at the consequences of transcortin deficiency on the regulation of the hypothalamic-pituitary-adrenal axis leading to glucocorticoid release. Although the free fraction of corticosterone is found markedly elevated in Cbg-/- (20-26%) and moderately in Cbg+/- (9-12%) compared to Cbg+/+ (7-9%), the concentration of free corticosterone in resting conditions is found unchanged between groups because total corticosterone levels are down-regulated particularly in Cbg-/- at the peak of secretion. However, in stress conditions this adjustment does not occur.

We have measured total and free corticosterone after a restraint stress and we show that the expected rise of total and free corticosterone is suboptimal in transcortin deficient mice (both Cbg+/- and Cbg-/-). At first sight this may seem surprising as a deficiency in transcortin would be expected to lead to increased free hormone concentration. In fact, the increased clearance from plasma of secreted corticosterone outweighs the elevated free fraction of corticosterone in transcortin mutant mice. Thus, deficiency in transcortin results in insufficient corticosterone levels after stress.

Then, we examine the impact of transcortin deficiency on behavioural stress responses. We found that Cbg+/- and Cbg-/- mice show altered response in two tests measuring despair-related behaviours: the forced swim and the learned helplessness tests. As for the restraint stress, the free corticosterone levels were ~30% lower in Cbg mutant mice after the behavioural tests. Furthermore the gene Egr-1 (also known as zif268, NGFI-A, or Krox 24, a downstream gene of the MAP kinase pathway involved in behavioral effect of glucocorticoids in brain) was under-expressed in the hippocampus of the same mice, suggesting glucocorticoid hyposignaling in brain. This idea was reinforced by an experience of cocaine sensitisation, known to be glucocorticoid levels dependent. The Cbg+/- present a weak sensitisation to repeated injections of cocaine whereas Cbg-/- mice show no sensitisation at all. Thus, we gathered converging data suggesting that transcortin deficient mice present a glucocorticoid hyposignaling after stress associated with altered behavioural responses.
Now we want to extend these results to learning and memory processes and to examine if corticosterone replacement will restore normal behaviour in the mutant mice. Additionally, the role of cerebral cytokines in the depressive-like behaviour will be studied as they will be up-regulated in a situation of glucocorticoid hyposignaling. Finally, we aim to study the role of transcortin gene polymorphisms in stress vulnerability in human cohorts.