TLR activation of the transcription factor XBP1 regulates innate immune responses in macrophages

Abstract

Sensors of pathogens, such as Toll-like receptors (TLRs), detect microbes to activate transcriptional programs that orchestrate adaptive responses to specific insults. Here we report that TLR4 and TLR2 specifically activated the endoplasmic reticulum (ER) stress sensor kinase IRE1alpha and its downstream target, the transcription factor XBP1. Previously described ER-stress target genes of XBP1 were not induced by TLR signaling. Instead, TLR-activated XBP1 was required for optimal and sustained production of proinflammatory cytokines in macrophages. Consistent with that finding, activation of IRE1alpha by ER stress acted in synergy with TLR activation for cytokine production. Moreover, XBP1 deficiency resulted in a much greater bacterial burden in mice infected with the TLR2-activating human intracellular pathogen Francisella tularensis. Our findings identify an unsuspected critical function for XBP1 in mammalian host defenses.

Figure 1. TLRs activate XBP1 mRNA maturation in the absence of ER-stress

(a), IRE1α activation (phos-tag SDS-PAGE), PERK phosphorylation and ATF6α processing in LPS or tunicamycin (TM) stimulated J774 cells extracts. (b), Induction of CHOP, BiP, PDI and Erdj4 mRNA was measured in untreated (6h), TM-treated (2, 4 and 6 h), TM plus LPS (grey bars), or Pam₃CSK₄ (black bars) treated J774 cells by real-time PCR relative to β-actin (mean and s.e.m. of triplicates). J774 cells were also left untreated (white bars). (c), Immunoblot of XBP1s, CHOP and processed ATF6α in nuclear extracts (Nuclear xt) of J774 macrophages stimulated with LPS or TM for the indicated times. Pro-IL-1β levels in total extracts (Total xt) is shown. XBP1 mRNA maturation was analyzed by RT-PCR (bottom). (d), The ability of muramyl dipeptide (MDP) and a panel of specific TLRs agonists to promote XBP1 activation with TM as a positive control was tested. XBP1 splicing and control genes IL-1β and IFN-β were monitored by RT-PCR. (e), XBP1 splicing by RT-PCR in LPS, Pam₃CSK₄ or TM treated J774 cells stably transduced with IRE1α or control shRNA lentiviruses. (f) XBP1 mRNA maturation was analyzed by RT-PCR in bone marrow derived macrophages (BMMs) from TLR4 competent (C3H/HeOuJ) or defective (C3H/HeJ) mice stimulated with LPS 100 ng/ml or 10 μg/ml TM (time and dose dependent concentrations of LPS indicated above lanes)., XBP1u, unspliced XBP1; XBP1s, spliced (mature) XBP1; p-IRE1, p-PERK, phosphorylated IRE1 and PERK respectively; ATF6p, processed form of ATF6α. Data are representative of at least three independent experiments.

Figure 2. XBP1 activation requires proximal TLR signaling

(a) BMMs from MyD88, TIRAP, TRIF (Ticam1) deficient mice or littermate controls, were stimulated with 100 ng/ml of the TLR4 agonist LPS, the TLR1/2 agonist Pam₃CSK₄, the TLR6/2 agonist FSL1 or 10 μg/ml TM as indicated. XBP1 mRNA maturation was analyzed by RT-PCR. (b,c) J774 cells stably transduced with lentivirus encoding TRAF6, and NEMO specific shRNA or a control shRNA (b), or pretreated for 5 min with NADPH oxidase inhibitors diphenyleneiodonium chloride (DPI, 10 μM), apocynin (1 mM) or the carrier DMSO (1 μl/ml) (c), were stimulated with LPS, Pam₃CSK₄ or TM as indicated and analyzed for XBP1 mRNA splicing by RT-PCR. To confirm knockdown efficiency, IL-6 production upon stimulation with LPS was measured by ELISA (b, bottom). (d), WT or NOX2-deficient macrophages (Cybb ^−/−) were stimulated as indicated and monitored for XBP1 activation. Data are representative of three independent experiments.

Figure 3. XBP1 is required for optimal TLR responses

(a-c) BMMs from XBP1 deficient (XBP1^Δ) or littermate (WT) mice were stimulated with LPS, Pam₃CSK₄, or FSL1 as indicated. Supernatants were collected and analyzed for the production of IL-6 (a). mRNA was harvested and analyzed by real-time PCR to quantify cytokines (b,c). A time course is shown in c and 6 h stimulation time points are shown in a and b. Relative expression to β-actin is given. Data are from one representative of four (a) or three (b, c) independent experiments (mean and s.e.m). ** P ≤ 0.01; * P < 0.02. ND, not detected.

Figure 3. XBP1 is required for optimal TLR responses

(a-c) BMMs from XBP1 deficient (XBP1^Δ) or littermate (WT) mice were stimulated with LPS, Pam₃CSK₄, or FSL1 as indicated. Supernatants were collected and analyzed for the production of IL-6 (a). mRNA was harvested and analyzed by real-time PCR to quantify cytokines (b,c). A time course is shown in c and 6 h stimulation time points are shown in a and b. Relative expression to β-actin is given. Data are from one representative of four (a) or three (b, c) independent experiments (mean and s.e.m). ** P ≤ 0.01; * P < 0.02. ND, not detected.

Fig. 4. IRE1α activation synergizes with TLRs to augment cytokine production

(a), Synergy for IL-6 production (ELISA) and ISG15 (RT-PCR) in J774 cells treated with carrier DMSO (black line) or 1 μg/ml TM (grey line) for 6 h in the presence or absence of LPS. Relative expression to β-actin is given. (b), J774 cells transduced with 1 or 10 μg of recombinant active IRE1α protein in the presence of endoporter (endo) for 1 h were stimulated with LPS for 6 h as indicated. Protein internalization and aggregation was monitored by immunoblot and Il6 mRNA production by quantitative RT-PCR. Relative expression to β-actin is given. (c), J774 cells stably transduced with IRE1α or control shRNA lentiviruses were stimulated with LPS for 6 h and TM as indicated. IL-6 mRNA was monitored by RT-PCR. Relative expression to β-actin is given. (d), XBP1^Δ or littermate (WT) BMM untreated or prestimulated for 16 h with 100 ng/ml TM and then incubated with LPS for 3 h in fresh medium. IL-6 production was monitored by RT PCR. (e) Cytokine production (RT-PCR) in human macrophages untreated (white bars) or treated with TM (light grey bars), infected with F. tularensis live vaccine strain (LVS) (grey bars) or infected with LVS in the presence of TM (black bars) for 3 h, 6 h and 18 h as indicated. Relative expression to untreated samples is given. Results are representative of at least two independent experiments (mean and s.d. of triplicate assays (c,d) or duplicate assays (e)).

Figure 5. XBP1 recruitment to Il6 and Tnf promoters

(a,b) ChIP analysis of XBP1 occupancy at the Il6 promoter in unstimulated or TM-stimulated macrophages derived from littermate (WT) or XBP1^Δ mice. ChIP assay using GST antibody was performed as negative control (Mock ChIP). Fold enrichment is the relative abundance of DNA fragments at the Il6 promoter (a) or at the Tnf promoter and enhancer (b) over a control region in the Apoa4 promoter as quantified by real-time PCR. Data are representative of three independent experiments, (mean and s.d) of triplicates).

Figure 6. XBP1 deficiency impairs resistance to F.tularensis LVS infection

(a) BMM were left untreated (−) or infected at MOI 1:10 with M. tuberculosis for 6 h, L. monocytogenes for 8 h or F.tularensis for 24 h as indicated. XBP1s was monitored by RT-PCR. (b) BMMs of Tlr2^−/−, Cybb^−/− and respective control mice were infected with F.tularensis at MOI 10:1, or MOI 100:1 for 8 h and analyzed for XBP1 splicing by RT-PCR. (c) BMMs from Tlr2^−/−, Cybb^−/−, XBP1^Δ, or littermate (WT) mice were infected with F. tularensis LVS at an MOI of 10:1. 8 h post-infection, IL-6, TNF and IL-1β production was analyzed by real-time PCR. (d) BMMs from Tlr2^−/−, XBP1^Δ or littermate (WT) mice were infected with F. tularensis LVS at an MOI of 10:1 for 8 h in the presence or absence of TM. Cytokine production was measured by real-time PCR. (e,f), XBP1^Δ or littermate (WT) mice were infected via the aerosol route. Nine mice per group (e) were monitored for bacterial counts in the indicated organs (CFU) 7 days post infection. Five mice per group (f) were monitored 14 days post infection. Data are represented with Box and Whiskers extended to extreme data points, P < 0.001. Data are representative of three independent experiments (a, c) and two independent experiments (b, d) (mean and s.e.m of triplicates (c, d).

Figure 6. XBP1 deficiency impairs resistance to F.tularensis LVS infection

(a) BMM were left untreated (−) or infected at MOI 1:10 with M. tuberculosis for 6 h, L. monocytogenes for 8 h or F.tularensis for 24 h as indicated. XBP1s was monitored by RT-PCR. (b) BMMs of Tlr2^−/−, Cybb^−/− and respective control mice were infected with F.tularensis at MOI 10:1, or MOI 100:1 for 8 h and analyzed for XBP1 splicing by RT-PCR. (c) BMMs from Tlr2^−/−, Cybb^−/−, XBP1^Δ, or littermate (WT) mice were infected with F. tularensis LVS at an MOI of 10:1. 8 h post-infection, IL-6, TNF and IL-1β production was analyzed by real-time PCR. (d) BMMs from Tlr2^−/−, XBP1^Δ or littermate (WT) mice were infected with F. tularensis LVS at an MOI of 10:1 for 8 h in the presence or absence of TM. Cytokine production was measured by real-time PCR. (e,f), XBP1^Δ or littermate (WT) mice were infected via the aerosol route. Nine mice per group (e) were monitored for bacterial counts in the indicated organs (CFU) 7 days post infection. Five mice per group (f) were monitored 14 days post infection. Data are represented with Box and Whiskers extended to extreme data points, P < 0.001. Data are representative of three independent experiments (a, c) and two independent experiments (b, d) (mean and s.e.m of triplicates (c, d).