Vasoconstrictor response to topical beclomethasone in major depression

Abstract

Overactivity of the hypothalamic–pituitary–adrenal (HPA) axis has been frequently described in depression. Due to the closed-loop nature of the HPA axis, one possible cause of this overactivity may be a defect in negative feedback regulation, in particular an abnormality of the glucocorticoid receptor (GR). In the present study, the vasoconstrictor response to the topical glucocorticoid, beclomethasone, was used to examine GR function in depression. Topical beclomethasone was applied in four concentrations (10 μl each of 3, 10, 30 and 100 μg/ml) to the forearms of 22 subjects with major depression and their age- and sex-matched controls. Skin blanching responses were compared between the depressed and control groups and, within the depressed group, on the basis of the modified dexamethasone suppression test (DST), between cortisol suppressors and non-suppressors. Depressed subjects demonstrated a significantly reduced vasoconstrictor response compared to controls (P=0.0001). No difference was detected between cortisol suppressors and non-suppressors in their skin blanching responses. These findings suggest that peripheral GR function is abnormal in depression but that the reduced vasoconstrictor response to beclomethasone is not necessarily a secondary effect of hypercortisolaemia or HPA axis overactivity.

Introduction

Numerous neuroendocrine abnormalities have been reported in depression that suggest overactivity of the hypothalamic–pituitary–adrenal (HPA) axis (Carroll et al., 1976). These include hypercortisolism (Sachar et al., 1973, Halbreich et al., 1985), elevated corticotropin-releasing hormone in the cerebrospinal fluid (Nemeroff et al., 1984) and failure to suppress cortisol after dexamethasone (Carroll et al., 1981). As the HPA axis operates as a closed-loop system (Houk, 1988), being regulated by feed-forward and feedback limbs, overactivity of the axis may be due to increased central drive, impaired negative feedback or a combination of both.

Negative feedback inhibition of the HPA axis is mediated by two types of corticosteroid receptors in the brain and pituitary: the mineralocorticoid receptor (MR) and the glucocorticoid receptor (GR). These receptors differ in anatomical distribution and functional activity, operating as a two-level recognition system for circadian and stress-induced changes in corticosteroids (Reul and De Kloet, 1985, Reul and De Kloet, 1986). MRs are primarily septo-hippocampal in distribution and have a high affinity but low capacity for cortisol. They are relatively insensitive to the dynamic variations in cortisol levels. GRs have a wider distribution and are present in the pituitary and many brain areas, including those thought to be involved in the stress response (Reul et al., 1987). GRs have a low affinity but high capacity for cortisol and are very responsive to changes in cortisol concentrations. Whereas MRs are thought to be involved in the tonic inhibitory activity within the HPA axis, GRs appear to “switch off” cortisol production at times of stress (De Kloet and Reul, 1987).

One hypothesis suggests that the overactivity of the HPA axis in depression may be due to an abnormality of the GR at the limbic–hippocampal level (see Modell et al., 1997). This abnormality results in a defect in or resistance to glucocorticoid negative feedback which, by causing inadequate restraint of the forward drive, manifests as HPA axis overactivity. Several findings in depression are consistent with an abnormality of the GR. Non-suppression of cortisol after dexamethasone, reported in up to 70% of patients with depression (APA Task Force, 1987), suggests a malfunctioning negative feedback that may be due to pituitary GR insensitivity (Halbreich et al., 1985, Jaeckle et al., 1987). This is supported by the finding that the individual sensitivity to dexamethasone in depressed non-suppressors is approximately half that of cortisol suppressors (Gupta et al., 1992). Patients with depression fail to show the physical stigmata of corticosteroid excess, despite the frequent presence of hypercortisolism (Murphy, 1991), suggesting that peripheral GRs may be abnormal or insensitive in depression. Studies of peripheral lymphocytes in depression also suggest an abnormality of the GR, with findings of reduced GR numbers (Gormley et al., 1985, Whalley et al., 1986, Yehuda et al., 1993) or abnormal GR functional activity (Schlechte and Sherman, 1985, Wassef et al., 1990, Rupprecht et al., 1991).

Corticosteroids influence vascular tone (Brenner et al., 1989) and cause blanching of the skin when topically applied. This cutaneous vasoconstrictor response has been used to compare the potency of topical steroids, as a high degree of blanching reflects high anti-inflammatory potency (McKenzie and Stoughton, 1962, Barry and Woodford, 1978). Vascular smooth muscle contains both GRs and MRs (Scott et al., 1987) but only the GR appears to be involved in cutaneous vasoconstriction. Skin blanching occurs with topical glucocorticoids, not mineralocorticoids, and is blocked by the administration of GR antagonists (Marks et al., 1982, Gaillard et al., 1985). The potent glucocorticoids, beclomethasone and budesonide, cause a high degree of skin blanching for a given concentration (Johansson et al., 1982) and have high affinities for the GR (Dahlberg et al., 1984). As skin blanching thus appears to be mediated by activation of the GR, the vasoconstrictor response to topical steroids provides a convenient probe of GR function.

In the present study, we examined the vasoconstrictor response to beclomethasone in subjects with depression. Skin blanching was compared between depressed subjects and healthy controls and then, within the depressed group after administration of the modified DST (Carroll et al., 1981), between cortisol suppressors and non-suppressors. Non-suppression of cortisol after dexamethasone was considered to reflect corticosteroid-resistance and provide an indicator of HPA axis overactivity. It was postulated that depressed subjects would have a reduced vasoconstrictor response compared to controls and that, within the depressed group, those with non-suppression of cortisol after dexamethasone or corticosteroid-resistance would have an even lesser response than those depressed subjects with corticosteroid-sensitivity. Should this be demonstrated, it would hopefully provide further evidence of defective GR function in depression, support the studies of peripheral lymphocyte glucocorticoid receptor abnormalities in depression and explore the relationship between HPA axis overactivity and receptor response.