- Top of page
- Materials and methods
Leptin regulates energy balance by suppressing appetite and increasing energy expenditure through actions in the hypothalamus. Recently we demonstrated that the effects of leptin are, at least in part, mediated by the release of interleukin (IL)-1β in the brain. Microglia constitute the major source of IL-1β in the brain but it is not known whether these cells express leptin receptors, or respond to leptin to produce IL-1β. Using RT-PCR and immunocytochemistry, we demonstrate that primary rat microglial cells express the short (non-signalling) and long (signalling) isoforms of the leptin receptors (Ob-R)s. Immunoassays performed on cell medium collected 24 h after leptin treatment (0.01–10 μg/mL) demonstrated a dose-dependent production and release of IL-1β and its endogenously occurring receptor antagonist IL-1RA. In addition leptin-induced IL-1β release occurs via a signal transducer and activator of transcription 3 (STAT3)-dependent mechanism. Western blot analysis demonstrated that leptin induced the synthesis of pro-IL-1β in microglial cells and the release of mature 17 kDa isoform into the culture medium. Leptin-induced IL-1β release was neither inhibited by the pan-caspase inhibitor BOC-D-FMK, nor by the caspase 1 inhibitor Ac-YVAD-CHO indicating that IL-1 cleavage is independent of caspase activity. These results confirm our earlier observations in vivo and demonstrate that microglia are an important source of IL-1β in the brain in response to leptin.
Interleukin-1 is a major mediator of inflammation, exerting a wide range of effects on the immune, endocrine and central nervous systems (CNS). IL-1β, the main released form of IL-1, exists as an inactive precursor molecule which requires cleavage by the enzyme caspase 1 into its biologically active ‘mature’ form. All actions of IL-1 are inhibited by a naturally occurring receptor antagonist (IL-1RA), which blocks IL-1 binding to its signalling receptor (Dinarello 1997). The activation of IL-1 signalling in the brain is an important regulator of systemic host defense responses to infection and inflammation including suppression of food intake and fever (e.g. increased thermogenesis) (Horai et al. 1998; Josephs et al. 2000).
Under normal, physiological conditions, brain IL-1 levels are extremely low and in most cases undetectable (Vitkovic et al. 2000), suggesting that this cytokine contributes little to physiological functions regulated by the brain. Recent evidence, however, suggests that, despite its low-expression, IL-1 could play a role in the homeostatic regulation of body weight and/or fat metabolism. IL-1RA deficient mice exhibit a lean phenotype and are resistant to diet-induced obesity when compared with their wild type controls, presumably due to enhanced/unchecked activity of IL-1 in the absence of the antagonist (Irikura et al. 2002; Matsuki et al. 2003; Somm et al. 2005). Conversely, mice lacking the IL-1R1 gene develop mature onset obesity (Garcia et al. 2006). Our own studies (Luheshi et al. 1999) and those of others (Garcia et al. 2006) demonstrated that the same mice (IL-1R1 knockouts) are resistant to the appetite suppressing effects of leptin. We also showed that administration of IL-1RA into the brain of normal rats abolishes the anorexic effect of leptin (Luheshi et al. 1999). We further found that neutralization of endogenous leptin by anti-leptin antiserum attenuated the increase of IL-1β mRNA expression in the hypothalamus, which was accompanied by reversal of the anorexia resulting from systemic inflammation induced by bacterial lipopolysaccharide (LPS) (Sachot et al. 2004). Collectively, these data indicate that the interaction between leptin and the IL-1 system in the brain plays an important role in body weight homeostasis both under physiological and inflammatory conditions. However, specific brain targets and cellular mechanisms of this interaction are largely unknown. As leptin exerts profound effects on peripheral immune cells, and microglia are the key immune cells in the CNS, we hypothesized that microglial cells are important target of leptin in the brain. Here we show that leptin induces the production and release of mature IL-1β and IL-1RA proteins in primary culture of rat microglia, and that the maturation of IL-1β by leptin is caspase 1 independent via an as yet unidentified mechanisms.
- Top of page
- Materials and methods
Earlier observations have demonstrated an important role of brain IL-1β in the anorexic, febrile and thermogenic effects of leptin (Luheshi et al. 1999; Sachot et al. 2004; Wisse et al. 2004; Garcia et al. 2006; Harden et al. 2006). The present study shows that microglial cells highly express both isoforms of leptin receptors, Ob-Ra and Ob-Rb, compared with astrocytes or neurones (Fig. 1), and demonstrate that leptin can activate microglia to produce mature IL-1β (Figs 2a and 4). These results add considerably to our understanding of the cellular actions of leptin in the CNS and suggest that the control of food intake and body weight by circulating leptin occurs at least partly via the production of IL-1β from microglial cells of the hypothalamus.
In contrast to undetectable amounts of IL-1β in the medium of untreated microglia, we detected considerable concentrations of its receptor antagonist IL-1RA (Fig. 2b). This observation is in agreement with published studies showing that IL-1RA is readily detectable in both the circulation and brain of normal human and rodents (Palin et al. 2004; Somm et al. 2005), as well as culture medium of untreated primary human liver cells (Gabay et al. 1997). However, one cannot exclude the possibility that microglia in primary cultures are partially activated resulting in high-levels of IL-1RA production. The treatment of microglia with leptin further induced the synthesis and release of IL-1RA over basal levels (Fig. 2b), demonstrating that leptin regulates the production of IL-1RA in parallel with that of IL-1β in microglia. The present in vitro data are in agreement with previous in vivo reports by Hosoi et al. (Hosoi et al. 2002a,b) demonstrating that systemic leptin injection increased mRNA expression of both IL-1β and IL-1RA. Because the balance between IL-1 and IL-1RA will determine the net IL-1 signalling, further investigation is required to characterize the precise role of leptin over the IL-1/IL-1RA ratio and its physiological significance. A recent study showed that serum IL-1RA concentrations are several fold higher in obese individuals, and they decreased after weight-loss (Meier et al. 2002), suggesting a possible implication of IL-1RA in the mechanism of leptin resistance. A separate study revealed that white adipose tissue constitutes a major source of IL-1RA under physiological conditions, and that IL-1RA production increases in obesity as well as in response to inflammatory stimuli (Juge-Aubry et al. 2003). These data combined with the fact that IL-1RA deficient mice show decreased fat mass and are resistant to high-fat diet-induced obesity (Somm et al. 2005), implicate IL-1RA in energy balance regulation.
Although leptin induces the expression and release of IL-1β from microglia, we found that the levels of IL-1β produced were relatively low when compared with those triggered by LPS, which is known to induce robust microglial activation. Whilst it is possible that a higher dose of leptin could induce an amount of IL-1β similar to that induced by LPS, it is plausible that the difference in the levels of this cytokine reflects an alternative physiological role for microglia that is distinct from that normally associated with inflammatory responses to exogenous pathogens or injury. CNS injuries, for example, trigger microglial activation leading to the production of large (pathophysiological) amounts of IL-1 (Allan and Pinteaux 2003). In contrast, leptin may trigger (or maintain) mild microglial activation, which could result in the production of smaller amounts of IL-1β. This low-concentration of IL-1β could contribute to the maintenance of normal body weight as demonstrated by a recent in vivo study by Garcia et al. (2006) using IL-1R1 knockout mice. Furthermore, hypothalamic expression of IL-1β is significantly reduced in response to decreased levels of leptin induced by acute starvation, or in Zucker rats, characterized by dysfunctional leptin receptors (Wisse et al. 2004). These data indicate a role for IL-1β as one of the downstream signals of leptin under physiological conditions.
The hypothesis that microglia and IL-1β may play a physiological role in body weight homeostasis under the influence of leptin, however, does not rule out inflammatory actions of leptin in the brain under pathophysiological conditions. We have shown previously in vivo that administration of leptin in rats induces fever (Luheshi et al. 1999), and that neutralization of endogenous leptin with anti-leptin antiserum attenuates LPS-induced fever (Sachot et al. 2004). More recently, we also reported that leptin induces cyclooxygenase-2 in the brain partly via IL-1 action (Inoue et al. 2006). Interestingly, Sanna et al. (2003) have reported that, in mice autoimmune encephalomyelitis, infiltrating T cells and macrophages produce leptin within the brain, suggesting a role of leptin for the development of certain neuroinflammatory diseases. Although some observations made in the current study would support this notion, further investigations are required to clarify the exact mechanisms that regulate the physiological and pathophysiological actions of this hormone.
Leptin regulates body weight through the activation of STAT3 (Bates et al. 2003), an intracellular signalling molecule activated by Ob-Rb. In the present study, we show that leptin activates STAT3, and demonstrate that inhibition of STAT3 significantly suppressed the leptin-induced IL-1β release in microglial culture (Fig. 3). This finding is somewhat contradictory to previous observations in vivo showing that leptin increased IL-1β mRNA expression in the hypothalamus of obese db/db mice which lack Ob-Rb/STAT3 signalling (Hosoi et al. 2002b). The reason for this discrepancy is unclear, but the different experimental conditions (mRNA levels in mouse brain tissue versus protein levels in microglial culture from rats) can be one possible explanation.
Biological activation of the IL-1β protein depends on proteolytic cleavage of the inactive 32 kDa precursor into a 17 kDa mature form, a mechanism mediated by caspase 1 (Kuida et al. 1995; Li et al. 1995). In the present study, we found that leptin-induced IL-1β release was not inhibited by a broad spectrum pan-caspase inhibitor (BOC-D-FMK) (Fig. 4a and b). Similar results were obtained using a different caspase inhibitor (Ac-YVAD-CHO), which effectively inhibited ATP-induced release of IL-1β from LPS-primed cells (Fig. 4c), a prototypical IL-1β release mechanism mediated by caspase 1. In contrast to its effect on the release of mature IL-1β from LPS-primed cells, ATP failed to induce the release of mature IL-1β from leptin-treated microglia (Fig. 5). These results collectively indicate that leptin-induced IL-1β release occurs independently of caspase 1 activity, and thus involves different mechanisms from LPS-induced IL-1β release. caspase 1-independent processing of IL-1β has already been reported (Miwa et al. 1998) and other extracellular proteases have been proposed for an alternative mechanism of IL-1β cleavage (Schonbeck et al. 1998; Herzog et al. 2005). However, whether or not these mechanisms are involved in the case of leptin was not addressed in the current study. Processing of 32 kDa pro-IL-1β by caspase 1 is thought to lead to the production of a 27 kDa intermediate form, which allows exposure of the cleavage site at Asp116 rendering it accessible to caspase 1 for full processing to the 17 kDa isoform. Western blot analysis showed the presence of an additional band of 14 kDa in the medium of leptin treated cells. An additional 14 kDa isoform of IL-1β has already been reported (Knudsen et al. 1986) and could be the product from direct cleavage of the 32 to the 17 kDa isoform.
Although our data clearly demonstrate that leptin interacts directly with microglial cells in vitro, how this interaction might occur in vivo is still an open question. Given the wide distribution of systemically injected radiolabelled leptin in the brain (Banks et al. 1996) and the abundant distribution of microglial cells throughout the CNS, it is feasible that once in the brain this hormone will activate microglia, regardless of location, resulting in IL-1β production (Hosoi et al. 2002b). However, it is also likely that the action of leptin on microglia occurs more neuroanatomically restricted to areas of the brain (for example the hypothalamus) as a result of a more restricted entry to specific regions such as those described for astrocytes acting as a delivery system for leptin to the arcuate nucleus of the hypothalamus (Cheunsuang and Morris 2005).
In summary, these observations suggest that microglia are a target for leptin action which leads to the production of the pro-inflammatory cytokine IL-1β. In combination with our previous findings in vivo (Luheshi et al. 1999), these results add further support to the hypothesis that leptin acts as a neuroimmune modulator, and suggest that microglial cells play an important part in this process.