In response to physiological and psychogenic stressors, the hypothalamic–pituitary–adrenal (HPA) axis orchestrates the systemic release of glucocorticoids (GCs). By virtue of nearly ubiquitous expression of the GC receptor and the multifaceted metabolic, cardiovascular, cognitive, and immunologic functions of GCs, this system plays an essential role in the response to stress and restoration of an homeostatic state. GCs act on almost all types of immune cells and were long recognized to perform salient immunosuppressive and anti-inflammatory functions through various genomic and non-genomic mechanisms. These renowned effects of the steroid hormone have been exploited in the clinic for the past 70 years and synthetic GC derivatives are commonly used for the therapy of various allergic, autoimmune, inflammatory, and hematological disorders. The role of the HPA axis and GCs in restraining immune responses across the organism is however still debated in light of accumulating evidence suggesting that GCs can also have both permissive and stimulatory effects on the immune system under specific conditions. Such paradoxical actions of GCs are particularly evident in the brain, where substantial data support either a beneficial or detrimental role of the steroid hormone. In this review, we examine the roles of GCs on the innate immune system with a particular focus on the CNS compartment. We also dissect the numerous molecular mechanisms through which GCs exert their effects and discuss the various parameters influencing the paradoxical immunomodulatory functions of GCs in the brain.
Any imbalances to an organism homeostasis elicit a complex stress response that involves the coordinated activation of functionally overlapping neuroendocrine and autonomic systems. Among these critical systems is the hypothalamic–pituitary–adrenal (HPA) axis, which is triggered by stressors of various sources (physical, emotional, immunological, etc.) to provoke the systemic release of glucocorticoids (GCs).
The activity of the HPA axis is regulated by multiple afferent sympathetic, parasympathetic, and limbic circuits (e.g., amygdala, hippocampus, and medial prefrontal cortex) innervating either directly or indirectly the paraventricular nucleus (PVN) of the hypothalamus. The PVN integrates converging stimulatory (catecholaminergic, glutamatergic, and serotonergic) or inhibitory (GABA-ergic) inputs, and thus represents a critical relay in the control of the HPA axis. The HPA axis is activated when secretory neurons of the medial parvocellular division of the PVN are stimulated, either directly or by relieving inhibitory inputs. As a result, corticotropin-releasing hormone (CRH) and arginine vasopressin (AVP) are both released in the portal circulation of the anterior pituitary gland. In turn, these neuropeptides trigger the secretion of adrenocorticotrophic hormone (ACTH) in the bloodstream by pituitary corticotrophs. ACTH then induces the production and the systemic release of GCs by the zona fasciculata of the adrenal cells.
Under basal conditions, the HPA axis exhibits a continuous oscillatory activity characterized by circadian and ultraradian variations. GCs are thus secreted in a highly pulsatile fashion throughout a 24 h cycle, displaying greater mean levels during the awake phase. The circadian rhythm of the HPA axis is orchestrated by the suprachiasmatic nucleus (SCN) of the hypothalamus and the oscillating release of GCs is believed to optimize stress responses. Interestingly, the ultraradian pulsatility of the HPA axis was recently associated with pulses of glucocorticoid receptor (GR)-mediated transcriptional regulation. Upon stress, the intensity and duration of the HPA response both depend on the specific nature of the encountered stressor. The precise circadian or ultraradian phase at which stress occurs also profoundly influences the systemic release of GCs, since higher levels are secreted when the challenge coincides with rising pulses.
Multiple non-exclusive pathways participate in the activation of the HPA axis upon cerebral or peripheral immune challenges. When an immunogenic insult takes place in the brain, various inflammatory mediators produced locally may trigger the HPA axis. In contrast, multiple routes convey stimulatory signals from the periphery to the HPA axis when a challenge occurs outside the CNS. In this scenario, circulating immunogenic or inflammatory factors may access and activate regulatory neuronal circuits (either directly or not) projecting to the PVN via the fenestrated endothelium of the circumventricular organs (CVOs) or a disrupted blood–brain barrier (BBB). Alternatively, circulating immune ligands may also bind their cognate receptor(s) anchored in the luminal membrane of endothelial cells of brain capillaries. Hence, they can simultaneously engage numerous transduction signaling pathways that disseminate the activating cues to the HPA axis through the parenchymal release of diverse inflammatory messengers. To this end, distinct lines of evidence support a critical role of MyD88, COX-2, microsomal prostaglandin E synthase (mPGES-1), and prostaglandin E 2 (PGE 2) in relaying peripheral stimulatory signals to the HPA axis. Although the exact cell types at play are still actively debated, endothelial cells are widely acknowledged as pivotal in this activating cascade. The contribution of perivascular cells (PVCs) however remains a controversial topic since they were identified either as a negligible or substantial source of PGE 2 in murine models of systemic inflammation. In addition to COX-2, more recent findings also support a role for COX-1 in activating the HPA axis. Finally, afferent fibers of the vagus nerve may also signal peripheral inflammation to the brain and thereby activate the HPA response.
The precise regulation of the activity of the HPA axis is of the utmost importance since both an excessive or insufficient release of GCs entail severe detrimental metabolic and immunological effects. As a matter of fact, chronic exposure to GCs results in various adverse side effects such as osteoporosis, diabetes, hypertension, dyslipidemia, and even neurodegeneration. On the other hand, a deficient or blunted HPA axis is commonly observed in the clinic in a wide range of autoimmune and inflammatory diseases. Likewise, disrupting the HPA axis surgically (through adrenalectomy) or pharmacologically (with GR antagonists) compromises the survival of normally resistant mice to septic shock. The magnitude and duration of the HPA response is thus tightly controlled by autoregulatory feedback loops involving the adrenals, pituitary, PVN, and upstream corticolimbic structures such as the hippocampus, amygdala, and medial prefrontal cortex. As a result, the HPA response is terminated through the same neuronal circuitry that mediates its activation.
The anti-inflammatory and immunosuppressive actions of GCs were first unraveled by the pioneering work of Kendall, Reichstein, and Hench more than 70 years ago and were since exploited in the clinic to treat a plethora of allergic, autoimmune, inflammatory, and hematological disorders as well as for preventing allograft rejection. GCs exert immunomodulatory functions by acting on practically every immune cell type, by virtue of the nearly ubiquitous expression of the GR. The cell-specific actions of GCs, which underlie the long-recognized anti-inflammatory and immunosuppressive effects of this steroid hormone, are briefly highlighted below.
Glucocorticoids strongly influence the phenotype, survival, and functions of monocytes and macrophages. GCs have long been recognized to increase the phagocytic potential of these critical effectors cells and thereby stimulate the clearance of foreign antigens, pathogens, inflammatory cells, cellular debris, and other potentially harmful elements. The steroid hormone also suppresses immunostimulatory functions and efficiently abrogates the production of pro-inflammatory mediators (such as cytokines, chemokines, and reactive oxygen or nitrous species) through various synergistic genomic and non-genomic mechanisms. In doing so, GCs promote an anti-inflammatory phenotype and expand the migratory activity and survival of these myeloid cells.
Glucocorticoids perform similar key functions in dendritic cells (DCs). In addition to regulating the maturation, survival, and motility of these antigen-presenting cells, GCs also hamper their immunogenic functions. Indeed, the end product of the HPA axis restricts the capacity of DCs to stimulate T cells by preventing the up-regulation of various co-stimulatory molecules, such as MHCII, B7.2 (CD86), and CD40. GCs can also convert DCs to tolerogenic cells, which promote the production of regulatory T cells. This immunosuppressive effect critically relies on the GC-mediated expression of glucocorticoid-induce leucine zipper (GILZ), since this transcription factor appears as both necessary and sufficient for the induction of a tolerogenic state. Recent work also indicates that the co-repressor DC-SCRIPT participate to this conversion. Interestingly, GCs exert distinct actions in immature and mature DCs and a disparate expression of various isoforms of the GR was recently found to underlie these divergent effects.
Another salient outcome of GCs administration is neutrophilia. The steroid hormone expands the number of circulating neutrophils by increasing their egress from the bone-marrow to the bloodstream and by concomitantly hindering their transmigration to inflammatory sites by alleviating the expression of cell adhesion molecules. Paradoxically, GCs were also shown to promote or attenuate neutrophil apoptosis, respectively through Annexin A1 or Mcl-1 and XIAP.
In contrast to neutrophils, GCs reduce the level
