Front cover: Glial fibrillary acidic protein immunostaining of mouse cortical astrocytes, exposed to retinoic acid. (see pp. 320–332). Photo: by Dr Jörg Mei, Institut für Biologie II, RWTH Aachen University, Germany.
The canonical interleukin-1 (IL-1) family is composed of three closely related proteins that are products of different genes: two agonists, IL-1α and IL-1β, and a naturally occurring IL-1 receptor antagonist (IL-1ra) [for detailed review, please see Dinarello (1994, 1996, 1998, 2002)]. In humans, IL-1α, IL-1β, and IL-1ra are encoded by the IL1A, IL1B, and IL1-RN genes, respectively, all located on the long arm of chromosome 2 (Webb et al. 1986; Modi et al. 1988; Lafage et al. 1989; Lennard et al. 1992; Steinkasserer et al. 1992; Patterson et al. 1993). Several new IL-1 family members, including IL-18 (Okamura et al. 1995; Bazan et al. 1996) and IL-1F5-11 (Debets et al. 2001; Nicklin et al. 2002; Taylor et al. 2002; Towne et al. 2004; Schmitz et al. 2005) have been identified based on gene homology, gene location, and receptor binding [for review, see Dunn et al. (2001); Dinarello (2002)]. Of the new family members, the most extensively studied is IL-18, which signals via binding through IL-1 receptor 18, also known as IL-1 receptor-related protein (IL-1Rrp) (Parnet et al. 1996; Torigoe et al. 1997). Activation of this receptor leads to induction and release of several pro-inflammatory cytokines [e.g., IL-6, IL-8, tumor necrosis factor (TNF)α, IL-1β, interferon (IFN)γ] (Dinarello 2002; Gracie et al. 2003). Hence, IL-18 signaling may contribute to tissue destruction in several inflammatory disorders (Dinarello 2002). It may also contribute to neuropathology [for review, see Felderhoff-Mueser et al. (2005)]. IL-1F6, IL-1F8, and IL-1F9 signal via a newly described receptor designated IL-1Rrp2 (Lovenberg et al. 1996; Towne et al. 2004), leading to activation of signaling pathways (i.e., mitogen-activated protein kinase and nuclear factor (NF)-κB activation) and downstream effectors (i.e., IL-6 release) similar to those activated by IL-1 (Towne et al. 2004). Since these ligands and their receptor are predominantly expressed in the skin and airways, it has been proposed that they represent the first line of defense against entering pathogens (Debets et al. 2001; Towne et al. 2004). IL-1F11, also known as IL-33, signals via activation of the IL-33 receptor ST2, resulting in increased production of IL-5 and IL-13, as well as reduced IFNγ release (Schmitz et al. 2005). Signaling pathways for IL-1F5 and IL-1F7 have yet to be elucidated. Interestingly, it appears that IL-1F5 resulted from a gene duplication of IL-1ra (Taylor et al. 2002). However, it does not appear to function as an IL-1RI antagonist, since it neither blocks IL-1α nor IL-1β-induced IL-6 production (Towne et al. 2004). Whether these new family members contribute to the pathogenesis or resolution of neurological disorders remains to be fully defined. Hence, this review will focus on the conventional (i.e., original) members of the IL-1 family, highlighting their potential to influence – both positively and negatively – neuropathological processes.
IL-1α and IL-1β share ≈ 30% structural homology and are both produced as 33 kDa precursors (Giri et al. 1985). While each can be cleaved to a 17 kDa ‘mature’ form by intracellular or extracellular proteases, most (≈ 90%) of IL-1α remains in the cytosol of cells in its precursor form or is transported to the cell surface where it remains membrane associated (Endres et al. 1989; Lonnemann et al. 1989). This membrane bound form may become activated and released following cleavage by an extracellular protease, perhaps now acting as a paracrine messenger to adjacent cells (Endres et al. 1989; Lonnemann et al. 1989; Dinarello and Wolff 1993). However, considering the intranuclear localization of IL-1α (Grenfell et al. 1989; Curtis et al. 1990), it has also been suggested that intracellular pro-IL-1α may directly function as a gene regulator (Maier et al. 1990; Kawaguchi et al. 2006). In contrast to IL-1α, proteolytic cleavage is required for biological activity of IL-1β whereby ≈ 80% of processed IL-1β– performed by the IL-1β-converting enzyme (ICE), also known as caspase-1 (Cerretti et al. 1992; Thornberry et al. 1992) – is released by the cell into the extracellular space (Dinarello 1996). Interestingly, IL-1 proteins, in general, lack the hydrophobic signal sequence (i.e., leader sequence) that targets most secreted proteins to the endoplasmic reticulum (Auron et al. 1984; Lomedico et al. 1984; March et al. 1985; Gray et al. 1986; Young and Sylvester 1989). Hence, it is unknown exactly how IL-1β is secreted, but it has been suggested that this can occur via exocytosis, active transport by a multi-drug resistance transporter, and/or following cell death (Hogquist et al. 1991; Griffiths et al. 1995; Singer et al. 1995; Ferrari et al. 1997; MacKenzie et al. 2001; Le Feuvre et al. 2002a,b; Andrei et al. 2004; Bianco et al. 2005; Brough and Rothwell 2007).
Despite being only 30% homologous, both mature IL-1α and IL-1β can exhibit an essentially identical repertoire of functions when presented exogenously (Dinarello and Thompson 1991; Dinarello 1998), although notable exceptions to this have been reported (Uehara et al. 1987; da Cunha et al. 1993; Juric and Carman-Krzan 2001). Binding to a specific 80 kDa plasma membrane receptor, designated the IL-1 receptor type I (IL-1RI) (Sims et al. 1993; Martin and Falk 1997; Loddick et al. 1998) facilitates the interaction with IL-1 receptor accessory protein (IL-1RAcP) (Wesche et al. 1997; Zetterstrom et al. 1998), which together induces downstream signaling pathways [for detailed review, see Sims and Dower (1994); Dinarello (1998); Martin and Wesche (2002); Li and Qin (2005)]. A second receptor designated IL-1 receptor type II (IL-1RII) can bind IL-1 but since it lacks the intracellular domain, it cannot signal. It functions biologically as a sink for IL-1β– it has a 10–100-fold lower affinity for IL-1α– and has been termed a decoy receptor (Colotta et al. 1993). Despite this, antibody blocking studies suggest that signaling through IL-1RII could mediate the febrile response as well as prostaglandin E2 release from hypothalamic explants elicited via exogenous administration of IL-1β (Luheshi et al. 1993; Mirtella et al. 1995). However, the specificity of this antibody to IL-1RII has been disputed (Gayle et al. 1994) and a very recent study using an endothelial-specific knockdown of IL-1RI calls the former observation into question (Ching et al. 2007). Nevertheless, whether an alternative signaling pathway (i.e., an as yet unidentified functional receptor for IL-1) exists remains unresolved (Desson and Ferguson 2003; Diem et al. 2003; Andre et al. 2006).
The third member of the IL-1 family, IL-1ra, serves to competitively inhibit IL-1α or IL-1β binding to IL-1RI and hence subsequent receptor signaling (Dripps et al. 1991; Dinarello 1998). However, it should be noted that high concentrations of IL-1ra relative to IL-1 are needed to block IL-1-mediated signal transduction [≈ 100-fold excess (Wakabayashi et al. 1991)] and an up-regulation of IL-1ra – as observed following CNS injury (Wang et al. 1997b) – may not be sufficient to counteract the effects of injury-induced IL-1 release. Additionally, the ligand binding portion of IL-1 receptors (both RI and RII) can be shed from the plasma membrane as soluble receptors where they are able to bind to circulating IL-1(α or β) with high avidity, thereby inhibiting the interaction between IL-1 and the cell surface IL-1RI (Roux-Lombard 1998). Given the potent biological activity of IL-1 – only a few receptors (in some cells < 10) need to be occupied for signaling to occur (Orencole and Dinarello 1989) – this regulation may be important to prevent initiation of potentially maladaptive inflammatory signaling processes.