Vagal Paraganglia Bind Biotinylated Interleukin-1 Receptor Antagonist: A Possible Mechanism for Immune-to-Brain Communication

Abstract

Interleukin-1β is a proinflammatory cytokine released by activated immune cells. In addition to orchestrating immune responses to infection, interleukin-1β is a key mediator of immune-to-brain communication. Interleukin-1β and endotoxin (which releases IL1β from immune cells) cause centrally mediated illness responses such as fever, aphagia, etc. These effects are blocked by intraperitoneal IL1 receptor antagonist (IL1ra), suggesting critical involvement of peripheral IL1 receptors. Centrally mediated illness responses are also blocked by vagotomy, suggesting that IL1β directly or indirectly activates vagal afferents. To test for IL1β binding, whole vagus (abdominal, laryngeal, and thoracic) and sections of hepatic vagus and liver hilus were incubated with biotinylated IL1ra and processed for avidin-biotin complex (ABC) or avidin-FITC histochemistry. Glomus cells of vagal paraganglia were labeled in all regions of the vagus. Biotinylated IL1ra also labeled smooth muscle and endothelial cells of blood vessels and lymphoid tissues. No label was present in omission or competition controls. These data suggest that centrally mediated illness responses result from IL1 activation of vagal paraganglia.

Introduction

Tissue damage and exposure to pathogens such as viruses or bacteria leads to activation of the immune system. As part of the normal immune response, activated macrophages release proinflammatory cytokines such as interleukin-1β (IL1β) [[18]]. IL1β is important in attracting, activating, and orchestrating the subsequent response of other immune cells [[39]]. In addition, IL1β is a key mediator of immune-to-brain communication [29, 36 43[60]]. Peripherally administered IL1β and endotoxin [lipo-polysaccharide (LPS); constituents of the outer cell walls of gram negative bacteria which induce release of IL1β from activated immune cells] cause a variety of “illness” responses mediated by the central nervous system, including fever, hyperalgesia, conditioned taste aversion, decreased food intake, decreased social interaction, alterations in brain monoamine levels, and activation of the hypothalamic-pituitary-adrenal axis [29, 36, 43, 60]. Illness responses produced by both IL1β and LPS are blocked by peripheral administration of recombinant human IL1 receptor antagonist (rhIL1ra) [43, 60], supporting the conclusion that specific IL1 receptors are critical to the effects observed.

Although the pathways by which IL1β signals the brain have not been conclusively identified (for review, see [[60]]), strong empirical support is mounting for an afferent pathway via peripheral nerves, most notably the vagus. Subdiaphragmatic vagotomy decreases the strength of conditioned taste aversions [[26]] and blocks fever [55, 59], hyperalgesia [61, 62], hypothalamic norepinephrine depletion [[22]], and elevated plasma corticosteroid levels [[22]] produced by peripheral LPS or IL1β. In addition, subdiaphragmatic vagotomy blocks IP LPS-induced decreased social interaction [[8]], decreased food intake [[9]], increased ACTH release [[25]], increased brain levels of mRNA for IL1 [[41]], and increased c-Fos expression in the brain [25, 58].

The effects of vagotomy appear to be selective for responses to peripheral proinflammatory cytokines, such as IL1β, mediated by the central nervous system. Vagotomy does not alter any of the following measures, compared to controls: peripheral immune responses [8, 41], brain c-Fos activation to electric shock [[58]], serum corticosteroid elevation to tailshock [[42]], alterations in positive or negative acute phase proteins to tailshock [[42]], generation of fever in response to centrally administered prostaglandin [[42]], or acquisition of conditioned taste aversions to intragastric copper or alcohol (see [[26]] for review). Such findings support the conclusion that a generalized suppression of central or peripheral responsivity cannot underlie the dramatic effects of subdiaphragmatic vagotomy on a host of illness responses. Rather, they strengthen the hypothesis that immune-to-brain communication is mediated by activation of afferent vagal nerves.

One mechanism by which this immune-to-brain communication could occur may be binding of IL1 to sensory elements of the vagus nerve. Of particular note are glomus cells of paraganglia, which form afferent synapses with the vagus [2, 3, 52]. Paraganglia are found in all major branches of the vagus nerve [40, 45, 51, 52]. Some paraganglia are embedded within the vagus nerve, whereas other, often larger, encapsulated paraganglia are situated apart from the main trunks of the vagus, and are connected to it via small fascicles of afferent nerves [2, 17]. Paraganglia consist primarily of glomus cells, although about 10% of abdominal paraganglia also contain a few neurons of unknown function [[2]]. Although the function or functions of paraganglia are unknown, they have been proposed to serve a chemoreceptive function [[45]] and so are well-suited as an intermediary between substances released by activated immune cells and afferent activation of the vagus.

In the present study, we used a biotinylated form of interleukin 1 receptor antagonist (bIL1ra), an endogenous cytokine that binds to IL1 receptors [[20]], as an anatomical tool to detect the presence of binding sites on sensory elements of the vagus. Biotinylation of an endogenous IL1 receptor antagonist provides a new method for examining the distribution of IL1 binding sites. Both types of IL1 receptors would be expected to be detected by this technique, because IL1ra binds both type I and type II receptors [[20]]. Because IL1ra is produced using recombinant technology (Amgen), sufficient quantities are available to allow biotinylation of rhIL1ra to be a practical method for anatomical studies. This new technique avoids the need to use iodine-125-labeled cytokines [38, 56, 57], and provides greater cellular resolution (e.g., enabling one to distinguish between labeled perineureum cells vs. nerve fibers) than either binding assays [38, 56, 57] or in situ hybridization [[15]]. This is particularly important when studying peripheral tissues, where a number of cell types express IL1 receptors that may be difficult to distinguish using autoradiographic techniques.