Of all the endocrine changes that are reported to occur in stress and depression, Cortisol hypersecretion is the most frequently observed. Although hypercortisolaemia can arise as a consequence of an acute stressor, there is a qualitative difference between the circadian pattern of Cortisol secretion in the depressed patient and that observed following exposure to a stressful stimulus. Thus the nadir of the plasma Cortisol concentration occurs some 6 hours earlier in the depressed patient than in the stressed, non-depressed patient. The ACTH concentration is also elevated while the concentration of corticotorphin releasing factor (CRF) in the cerebrospinal fluid is raised in depressed patients (Linkowski, Mendlewicz, and Kerklofs, 1987; Nemeroff, Widerlov, and Bissette, 1984). As there is a close association between plasma glucocorticoids and immune function, it is often assumed that the immune changes are a direct consequence of the raised plasma, and brain concentrations of the adrenal glucocorticoids. However, this seems unlikely as the prolonged increase in glucocorticoids in depression leads to a decrease in the sensitivity of the CRF and glucocorticoid receptors in the brain, pituitary, and on immune cells (Leake, Perry, and Ferrier, 1980; Dinan, 1994) thereby reducing the inhibitory effect to these steroid hormones on cellular immunity.
What causes the adrenal steroid abnormalities in depression? One possibility is that the elevated proinflammatory cytokine IL-1 is at least partly responsible. It is known that IL-1 has a direct action on the hypothalamus leading the increased release of CRF. In vitro, IL-1 has been shown to stimulate ACTH and growth hormone secretion (Goetzl, Screedharan, and Harkonen, 1988); these effects are not shared by the other major proinflammatory cytokines, IL-6, and TNF alpha. In addition to the direct effect of IL-1 on the HPA axis, there is controversial experimental evidence that macrophages may also secrete ACTH which could directly stimulate the adrenals to synthesize Cortisol. Thus an activated macrophage system in depression could both directly and indirectly contribute to hypercortisolaemia.
The question arises regarding the mechanism whereby an increase in peripheral IL-1 can precipitate changes in the immune, endocrine, and neurotransmitter systems in the brain. Indeed, an important and unresolved question is whether cytokines released peripherally gain access to the brain in concentrations that are biologically effective.
Cytokines are large hydrophilic molecules (Hamblin, 1994), their size and structure being such that passive diffusion across the blood brain barrier (BBB) is likely to be minimal (Hopkins and Rothwell, 1995). Currently, it is postulated that cytokines produced in the periphery can act on one or other circumventricular organs, such as the median eminence (ME) and the organum vasculosum laminae terminalis (OVLT) that lack a functional BBB (Hopkins and Rothwell, 1995). It has been suggested that cytokines from the periphery bind directly to glial cells on the OVLT, which in turn produce cytokines and other mediators such as prostaglandins, particularly prostaglandin E2 (PGE2). This hypothesis is consistent with the observations that peripheral IL-1 beta administration elevates PGE2 concentrations in many brain structures as assessed by in vivo microdialysis, which is maximal and most rapid at the OVLT and the medial preoptic area, and that the central increase in PGE2 precedes the onset of fever (Komaki, Arimura, and Koves, 1991). This hypothesis is further strengthened by the fact that many cytokine- and endotoxin-induced neurochemical (Lavicky and Dunn, 1995; Mefford and Heyes, 1990) and behavioural (Crestani, Seguy, and Dantzer, 1991; Hellerstein, Meydani, Meydani, Wu, and Dinarello, 1989; Osaka, Kannan, Kawano, Veta, and Yamashita, 1992) responses are attenuated by cyclooxygenase inhibitors such as indomethacin. Moreover, certain neurons in the preoptic nucleus have receptors for IL-1, IL-6, and TNF-alpha (Schettini, 1990). Finally, it has been reported in rodents that peripheral injection of an endotoxin increases the production of cytokines in the brain. Thus peripheral infections might affect the activity of the human brain at least in part through a similar mechanism (Rivier and Rivest, 1993).
Not only does IL-1 act at the OVLT, but it can cross the BBB by an active transport system (Banks, Kastin, and Durham, 1989). For instance, an active transport mechanism for TNF-alpha has been described (Gutierrez, Banks, and Kastin, 1993). The concentration of these cytokines crossing the BBB by such mechanisms may be so low that they are physiologically insignificant (Hopkins and Rothwell, 1995), but this may be an important route of entry into the brain when plasma concentrations of cytokines are very high (Banks et al., 1989).
In addition to the hypothesis that peripherally produced cytokines affect CNS function via the circumventricular organs, there is also evidence suggesting the existence of neurally mediated mechanisms of communication between peripherally produced cytokines and the CNS (Dantzer, 1994). This hypothesis that postulates communication between peripherally produced cytokines and the CNS via a neural afferent pathway is supported by the fact that subdiaphramatic vagotomy attenuates endotoxin induced depressive effects on behaviour, c-fos expression in the CNS and IL-1 beta expression in the hypothalamus (Dantzer, 1994). In addition, subdiaphramatic vagotomy has been reported to block HPA-axis activation produced by peripheral IL-1 beta and TNF-alpha administration (Fleshner, Bellgrau, Watkins, Laudenslager, and Maier, 1995) and also hypothalamic noradrenaline depletion produced by peripheral IL-1 beta administration. The findings from the vagotomy studies are important because they indicate that the brain is able to respond to cytokines that have been released at the periphery during the course of an infection or an inflammatory response and to respond to this stimulus by a local synthesis of cytokines. The mechanisms that are responsible for the transformation of the immune message into a neuronal message at the periphery, and the transduction of this neuronal message back into an immune message in the central nervous system, still need to be determined (Dantzer, Bluthe, Aubert, Goodall, Parnet, and Kelley, Bret-Dibat, Kent, Goujon, and Laye, 1996). Thus, in addition to cytokines produced from microglia and other macrophages within the CNS, peripherally produced cytokines can also affect the brain and produce many physiological, behavioural, endocrine, and neurochemical changes following most immunological challenges.
There is now substantial evidence to show that major depression is accompanied by an acute phase protein response, an increased secretion of prostaglandins and by an excessive secretion of proinflammatory cytokines. These and other changes suggest that immune activation may play a role in the pathogenesis of depression and provide the basis for the macrophage theory of depression (Smith, 1991). Thus inflammatory cytokines or lipopolysaccharide (LPS) administered to animals or man provoke an extensive set of symptoms that are also identical to those found in major depression. These changes affect not only the psychological state of the individual but are also associated with changes in the activity of the HPA similar to those seen in depression. There is also evidence that the increase in the circulatory concentrations of IL-1 and IL-6 mediate the acute phase protein response which is characterized by elevated positive acute phase proteins (Song, et al., 1994) and a reduction in the serum tryptophan concentration.
The precise mechanism whereby pro-inflammatory cytokines such as IL-1 beta modulate central serotonergic function is uncertain but recent evidence from in vitro studies on JAR cells, components of a choriocarcinoma cell line derived from human placenta, have shown that IL-1 activates the serotonin transporter directly (Ramamoorthy, Ramamoorthly, Prasad, Bhat, and Mahesh, 1995). If a similar effect occurs on central serotonergic neurons, it could lead to an increased removal of the amine from the synaptic cleft thereby leading to a reduction in serotonergic function. Receptors for IL-1 beta occur on serotonergic neurons (Cunningham and de Souza, 1996) and it is well established that this cytokine is synthesized by neurons and glial cells. Furthermore, raphe neurons may also respond to IL-1 delivered by white blood cells penetrating endothelial barriers during an inflammatory process (Cunningham and de Souza, 1996). Thus a reduction in the serum tryptophan concentration associated with elevated acute phase proteins, and an enhanced reuptake of serotonin from the synaptic cleft caused by the action of IL-1 on the serotonin transporter, may contribute to a malfunction of the serotonergic system that is causally associated with depression. Indirect support for this hypothesis comes from the observation that antidepressants suppress the pro-infammatory cytokines (Xia, De Piere, and Nassberger, 1996). This suggest that the proinflammatory cytokines can act as common mediators for the action of external (for example, psychosocial) and internal (for example, infections and toxins) stressors that are known to play a crucial role in the aetiology of depression.
One advantage of the macrophage theory of depression is that it brings together the disparate changes in the immune, endocrine, and neurotransmitter system with the clinical and epidemiological observations and also provides direct predictions that may be tested experimentally and/or clinically. For example, Maes, Smith, Christophe, Cosyns, Desnyder, and Meltzer (1996) have shown that the concentration of omega 3 fatty acids in the erythrocyte membranes of depressed patients is significantly decreased. This could suggest an inbalance between the omega 3 and omega 6 fatty acid pathway and reflects an increased synthesis of prostaglandins due to the relatively high intake of vegetable oil (a source of omega 6 fatty acids). Thus an increase in omega 6 and/or a decrease in omega 3, fatty acids could contribute to the changes that cause depression. A diet rich in omega 3 fatty acids (fish oil) might therefore have some immunoprotective function (Smith, 1991; Maes and Smith, 1998).
Was this article helpful?