LPS-induced release of IL-6 from glia modulates production of IL-1β in a JAK2-dependent manner

Abstract

Background: Compelling evidence has implicated neuroinflammation in the pathogenesis of a number of neurodegenerative conditions. Chronic activation of both astrocytes and microglia leads to excessive secretion of proinflammatory molecules such as TNF α, IL-6 and IL-1 β with potentially deleterious consequences for neuronal viability. Many signaling pathways involving the mitogen-activated protein kinases (MAPKs), nuclear factor κ B (NF κ B) complex and the Janus kinases (JAKs)/signal transducers and activators of transcription (STAT)-1 have been implicated in the secretion of proinflammatory cytokines from glia. We sought to identify signaling kinases responsible for cytokine production and to delineate the complex interactions which govern time-related responses to lipopolysaccharide (LPS).

Methods: We examined the time-related changes in certain signaling events and the release of proinflammatory cytokines from LPS-stimulated co-cultures of astrocytes and microglia isolated from neonatal rats.

Results: TNF α was detected in the supernatant approximately 1 to 2 hours after LPS treatment while IL-1 β and IL-6 were detected after 2 to 3 and 4 to 6 hours, respectively. Interestingly, activation of NF κ B signaling preceded release of all cytokines while phosphorylation of STAT1 was evident only after 2 hours, indicating that activation of JAK/STAT may be important in the up-regulation of IL-6 production. Additionally, incubation of glia with TNF α induced both phosphorylation of JAK2 and STAT1 and the interaction of JAK2 with the TNF α receptor (TNFR1). Co-treatment of glia with LPS and recombinant IL-6 protein attenuated the LPS-induced release of both TNF α and IL-1 β while potentiating the effect of LPS on suppressor of cytokine signaling (SOCS)3 expression and IL-10 release.

Conclusions: These data indicate that TNF α may regulate IL-6 production through activation of JAK/STAT signaling and that the subsequent production of IL-6 may impact on the release of TNF α, IL-1 β and IL-10.

Figure 1

LPS stimulates activation of JAK/STAT, c-jun and NFкB signaling pathways and release of proinflammatory cytokines from glial cells. Stimulation of glial cells with LPS (100 ng/ml) enhanced the expression of TNFα mRNA at 30 minutes (A; ^*P < 0.05;ANOVA; n = 3), release of TNFα at 1 h (B, see inset; ^*P < 0.05; Student t-test for independent means; n = 3), IL-1β mRNA expression at 1 h (C; ^**P < 0.01; ANOVA; n = 3), IL-1β release at 3 h (D; ^*P < 0.05; ANOVA; n = 3), IL-6 mRNA expression at 2 h (E; ^*P < 0.05; ANOVA; n = 3) and IL-6 release at 4 h (F; ^***P < 0.001; ANOVA; n = 3). Sample immunoblots (representative of three separate experiments) indicate that expression of phosphorylated-IκB, -c-jun, -JAK2 and STAT1 were all enhanced in glial cells incubated with LPS for 10, 30, 60 and 120 minutes (G). Expression of phosphorylated JAK1 was unchanged in cells that were treated with LPS (H).

Figure 2

Inhibition of JAK2 attenuates LPS-stimulated release of TNFα and IL-6 from glial cells. A sample immunoblot indicates that expression of phosphorylated JAK2 and STAT1 were enhanced in glial cells treated with LPS (2 h; 100 ng/ml), and that this effect was attenuated when cells were incubated in the presence of a specific JAK2 inhibitor, SAR317461 (2 μM) (A). Expression of phosphorylated IкBα was enhanced in cells that were treated with LPS (B); however, co-incubation of cells with LPS and SAR317461 had no effect on the LPS-induced expression of phosphorylated IкBα (B). The LPS-stimulated release of TNFα at 6 h (C; ^***P < 0.001; ANOVA; n = 3) was attenuated when cells were co-incubated with LPS and SAR317461 (⁺⁺P < 0.01; ANOVA; n = 3), whereas inhibition of the LPS-induced expression of phospho-JAK2/STAT1 using SAR317461 enhanced the LPS-induced release of IL-1β at 9 h (D; ⁺⁺⁺P < 0.001; ANOVA; n = 3). The LPS-induced up-regulation of IL-6 mRNA (E; ^***P < 0.001; ANOVA; n = 3) was attenuated when cells were co-treated with SAR317461 (E; ⁺⁺⁺P < 0.001; ANOVA; n = 3) as was the release of IL-6 (F; ⁺⁺⁺P < 0.001; ANOVA; n = 3).

Figure 3

JAK2 is phosphorylated and associates with TNFR1 in response to TNFα treatment. TNFα receptor (TNFR1) was immunoprecipitated from cell lysates of glia that had been treated with TNFα (5 ng/ml) for 0, 5 or 10 minutes, and western-blotted using a phospho-JAK2 specific antibody. Incubation of glial cells with TNFα stimulated a rapid association between phosphorylated JAK2 and TNFR1 (A). TNFα treatment of glial cells also induced phosphorylation of STAT1 at 10 minutes (B). The TNFα-stimulated phosphorylation of JAK2 and STAT1 was attenuated when cells were co-incubated with SAR317461 (C). Representative immunoblots of three separate experiments are shown. Incubation of glial cells with TNFα (5 ng/ml) stimulated release of IL-6 at 6 h (D, ^***P < 0.001; ANOVA; n = 3), an effect that was inhibited when cells were co-treated with TNFα and TG101209 (D, ⁺⁺⁺P < 0.001; ANOVA; n = 3). TNFα had no adverse effect on cell viability after 24 h of treatment (E, *P < 0.05; Student’s t-test for independent means; n = 3).

Figure 4

IL-6 modulates the LPS-stimulated release of IL-1β from glia. Glial cells were treated with LPS)(100 ng/ml) in the presence or absence of rat recombinant IL-6 (20 ng/ml) and cytokine concentrations were assessed at 24 h (A,C) and SOCS3 expression was examined at 90 minutes (B). The LPS-induced release of IL-1β (A; ^***P < 0.001; ANOVA) was partially inhibited when cells were co-incubated with recombinant IL-6 (A; ⁺P < 0.05; ANOVA). SOCS3 expression was induced in the presence of LPS (B) and further enhanced when cells were co-treated with LPS and recombinant IL-6 (B). IL-10 production was stimulated by LPS treatment (C; ^***P < 0.001; ANOVA), an effect that was greater when cells were stimulated with LPS in the presence of recombinant IL-6 (C; ⁺P < 0.05; ANOVA). Furthermore, inhibition of IL-6 signaling using a neutralizing antibody to the IL-6 receptor (IL-6R) exacerbated the LPS-induced effect on IL-1β release (D; ⁺⁺P < 0.01; ANOVA) compared to the IgG istoype control antibody while inhibiting the LPS-stimulated release of IL-10 (E; ⁺P < 0.05; ANOVA). Similarly, when the LPS-stimulated production of IL-6 was inhibited using siRNA targeted against IL-6 (G; ⁺⁺⁺P < 0.001; ANOVA), the LPS-induced release of IL-1β was enhanced (F; ⁺P < 0.05; ANOVA), effects that were not apparent in cells that were incubated in the presence of LPS and the non-target (NT) siRNA (F,G).

Figure 5

Schematic representation of LPS-induced signaling and cytokine release. Toll-like receptor (TLR)4 is stimulated by LPS and triggers activation of MAPKs and NFkB (1) which results in release of TNFα and IL-1β (2). Released TNFα interacts with its receptor (3), activating JAK2 and STAT1 (4) which up-regulates transcription and release of IL-6 (5). IL-6 feeds back to inhibit further release of TNFα and IL-1β (6), limiting the inflammatory changes.