J Physiol Volume 550, Number 1, 1-, July 1, 2003 DOI: 10.1113/jphysiol.2003.045609
J Physiol (2003), 550.1, p. 1
© Copyright 2003 The Physiological Society
DOI: 10.1113/jphysiol.2003.045609
Immune signalling to the brain
Quentin J Pittman and Abdeslam Mouihate
Neuroscience Research Group, Department of Physiology & Biophysics, University of Calgary, Calgary, Alberta, Canada
Email: pittman{at}ucalgary.ca
In the presence of bacteria and other infections, an innate immune response is generated to combat the invading pathogens. Peripheral immune cells such as macrophages, Kupffer cells and monocytes recognize components of the bacterial cell walls (e.g. lipopolysaccharide) or viral molecules via surface Toll-like receptors (TLRs). These cells synthesize and release a number of pro-inflammatory cytokines, among which are interleukin-1
(IL-1), interleukin-6 (IL-6) and tumour necrosis factor-
(TNF-).
The body responds to these molecules with an orchestrated series of physiological and behavioural responses. There is a profound activation of the central nervous system to cause fever, activate the hypothalamic- pituitary-adrenal axis and initiate a constellation of behavioural changes collectively called 'sickness behaviour' and which include fatigue, anorexia and social isolation. The means by which large circulating cytokine molecules signal the brain has preoccupied neuroimmunologists for many years, as these molecules are unlikely to pass the blood- brain barrier in biologically significant amounts. There is now consensus that a number of routes are possible, including activation of peripheral afferent nerves (e.g. vagus), direct binding to endothelial cells of the brain vasculature and activation of cells within the circumventricular organs (CVO) that lack a blood-brain barrier (Blatteis et al. 1998). In this issue of The Journal of Physiology, Desson & Ferguson (2003) have explored the action of IL-1 on the electrical activities of neurons of the subfornical organ (SFO), a highly vascularized forebrain CVO that is thought to be a critical interface for detecting plasma molecules and transducing this information into a neural signal. Desson & Ferguson have carried out a patch clamp study of isolated SFO neurons and report that physiological concentrations of IL-1 (100 fM to 1 pM) induced a transient depolarization of an identified subclass of SFO neurons that was brought about by reduction of a delayed rectifier potassium current and concurrent activation of a non-specific cation conductance. Interestingly, the very same cells also responded with a sequential hyperpolarization that was much more prominent at higher (septic) levels of IL-1. The IL-1 receptor antagonist (IL-1ra) blocked the depolarization, but not the hyperpolarization.
This study is the first to report electrophysiological actions of IL-1 at a CVO and furthermore is among the first to detect responses to IL-1 at the low levels seen in models of immune activation. However, the paper represents a beginning, not an end, as it raises numerous questions that are ripe for further investigation.
The first question pertains to the identity of the cells activated by the IL-1. Because the hypothalamic paraventricular nucleus (PVN) is one of the most highly activated structures after peripheral immune activation, and lesions of this structure interfere with fever development, one might have predicted that PVN-projecting SFO neurons would be most responsive to IL-1. However, in the study of Desson & Ferguson (2003), the SFO neurons that responded to IL-1 did not display the electrophysiological fingerprint characteristic of SFO neurons projecting to the PVN. With the use of tract tracing techniques, it should be possible to identify where the IL-1 responsive cells project; a likely site is the preoptic area, where neurons whose activities change during fever are known to exist.
A second question pertains to the signalling mechanism employed by IL-1 to alter the ionic conductances underlying the depolarization. The IL-1 receptor (IL-1R) exhibits many unique features (Martin & Wesche, 2002), at least in peripheral immune tissue and cell cultures, that are different from the G-protein linked and ionotropic receptors well studied by neurophysiologists. The IL-1R is part of the TLR superfamily and it is composed of an extracellular immunoglobulin-like domain and an intracellular signalling motif known as the Toll/IL-1 receptor (TIR) domain (the proliferation of abbreviations in the immunological literature is daunting to the novice!). For signal activation, IL-1 binds to the IL-1R, which is then recognized by an IL-1 receptor accessory protein (IL-1RAcP) to form a heterodimeric transmembrane complex. An intracellular docking molecule (MyD88) is recruited which anchors a series of IL-1R associated serine/threonine kinases (IRAK-1 to -4) that auto- or cross-phosphorylate to allow interaction with TNF receptor associated factor-6 (TRAF-6). This results in a series of phosphorylation steps that activate I
B kinases to release IB and permit NFB to translocate to the nucleus to act as a transcription factor. Whether this pathway is important in the electrophysiological changes reported by Desson & Ferguson (2003) is unknown, but the availability of transgenic mice with knockouts of various constituents of this complex pathway may provide an avenue for further investigation. In addition, a prominent consequence of the NFB activation is induction of enzymes important in prostaglandin synthesis (PGE synthase, cyclooxygenase-2) and nitric oxide generation (NOS). These molecules are critical components of the febrile response; however, the fact that inhibition of prostaglandin synthesis with cycoloxygenase inhibitors does not reduce FOS expression in the SFO suggests that alternate pathways may be important (Zhang & Rivest, 2000). A likely pathway is the mitogen activated protein kinase (MAPK) pathway, which is also rapidly activated by IL-1R occupancy and which has been implicated in some of the neural effects of IL-1 reported to date (Vereker et al. 2000).
In the study of Desson & Ferguson (2003), the hyperpolarization that occurred after higher doses of IL-1 was not further investigated. The fact that it was resistant to blockade by the IL-1ra suggests that it is not mediated by the IL-1R. However, the latter observation was only reported for two cells, suggesting that further data may be required to substantiate this claim. Whatever the receptor basis for the hyperpolarization, it is intriguing that the same cells can exhibit a sequential depolarization-hyperpolarization to IL-1. Thus, the same CNS pathways activated by an initial exposure to IL-1 may subsequently be inactivated, particularly at higher doses of the cytokine. One suspects that such effects may contribute to some of the hypotensive and hypothermic responses seen during sepsis.
These studies lay the groundwork for further studies on IL-1 action in other CVOs implicated in immune signalling to the brain and provide a useful starting point for an investigation of the cellular pathways utilized by IL-1 to acutely activate neurons. It will also be interesting to contrast the actions reported for IL-1 with that of the other major pro-inflammatory cytokines (TNF and IL-6) that appear in the circulation during fever.
 |
REFERENCES |
| Blatteis CM, Sehic E & Li S (1998). Ann N Y Acad Sci 856, 95-107. Desson SE & Ferguson AV, 2003 |
