It is well established that the hypothalamic-pituitary-adrenal axis (HPA) is activated by both external and internal stressors which result in the hypersecretion of adrenal glucocorticoids. In major depression the prolonged elevation of the glucocorticoid concentration leads to a desensitisation of the central glucocorticoid receptors and probably those receptors located on macrophages. These changes may account for the observation that many aspects of cellular immunity are activated in depression (for example, the increased release of pro-inflammatory cytokines from activated macrophages in the periphery and brain, and the increased release of acute phase proteins from the liver) even though other aspects of immunity (for example, natural killer cell activity and T-cell replication) are depressed. It is also known that some of the pro-inflammatory cytokines are potent activators of the HPA axis. Evidence is provided that the consequences of the hypersecretion of glucocorticoids and pro-inflammatory cytokines result in the malfunctioning of noradrenergic and serotonergic neurotransmission in the brain, changes which are reflected in the major symptoms of depression. Support for this view is provided by observations of the effects of some of these cytokines in non-depressed individuals being treated with pro-inflammatory and related cytokines for cancer. This has led to the hypothesis that depression is a form of sickness behaviour which forms the basis of the macrophage theory of depression. The review concludes with a discussion of the role of antidepressants in attenuating the adverse effects of glucocorticoids and pro-inflammatory cytokines on central neurotransmission. Although the precise mechanisms whereby antidepressants these changes is uncertain, there is evidence that they reduce the release of pro-inflammatory cytokines from activated macrophages and thereby facilitate the feedback inhibition of the HPA axis; this results in a reduction in the release of glucocorticoids from the adrenal glands. In addition, many antidepressants have been shown to increase the release of endogenous cytokine antagonists such as interleukin-1 receptor antagonist and interleukin-10. Evidence is also presented to show that different classes of antidepressants act as cyclooxygenase inhibitors which, by lowering the concentration of inflammatory prostaglandins in the brain, reduce the detrimental impact of the inflammatory changes on neurotransmitter function. An advantage of the macrophage hypothesis is that it extends the biogenic amine hypothesis of depression to take account of changes in the endocrine and immune systems which also play a crucial role in the aetiology of depression. In addition, the macrophage hypothesis may broaden the basis of understanding the mechanism of action of antidepressants.