Tpl2 transduces CD40 and TNF signals that activate ERK and regulates IgE induction by CD40

Abstract

Macrophages from Tpl2 knockout (Tpl2(-/-)) mice exhibit a defect in ERK activation by lipopolysaccharide (LPS). This impairs the nucleocytoplasmic transport of the tumor necrosis factor alpha (TNF-alpha) mRNA and prevents the induction of TNF-alpha by LPS. As a result, Tpl2(-/-) mice are resistant to LPS/D-galactosamine-induced shock. We demonstrate that Tpl2 is essential for ERK signals transduced by members of the TNF receptor superfamily, such as CD40 and the TNF receptor 1. Thus, ERK activation was impaired in Tpl2(-/-) B cells and macrophages stimulated with agonistic CD40 antibody or TNF-alpha, whereas the induction of other mitogen-activated protein kinases, such as JNK and p38, and the activation of NF-kappaB were unaffected. Tpl2 was recruited to a CD40/TRAF6 complex in response to CD40 stimulation. Moreover, TRAF6, which when overexpressed activates ERK, failed to do so in Tpl2(-/-) cells. The selective signaling defect resulting from the inactivation of Tpl2 allowed us to demonstrate that CD40-mediated ERK activation contributes to immunoglobulin production but is not essential for B-cell proliferation.

Fig. 1. ERK activation by CD40 in B cells and macrophages depends on signals transduced via Tpl2. (A) Splenic B cells from Tpl2^+/+ and Tpl2^–/– mice were stimulated with anti-CD40. A western blot of cell lysates harvested at the indicated time points before and after anti-CD40 stimulation was probed with an antibody that recognizes phosphorylated ERK (upper) or total ERK (lower). (B and C) Western blots of the same cell lysates were probed with antibodies against phosphorylated or total JNK (B) and phosphorylated or total p38 MAPKs (C). (D–F) Western blots of cell lysates derived from unstimulated and anti-CD40-stimulated bone marrow-derived macrophages were probed with antibodies against phosphorylated or total ERK (D), phosphorylated or total JNK (E) and phosphorylated or total p38 MAPKs (F). At least four independent experiments were performed and gave similar results.

Fig. 2. CD40 stimulation promotes comparable levels of NF-κB activation in B cells derived from Tpl2^+/+ and Tpl2^–/– mice, as determined by electrophoretic mobility shift assay (lanes 1–6). The specificity of the NF-κB binding was confirmed by the effects of an excess of cold NF-κB versus CREB-oligo probe (lanes 7 and 8), and the composition of the DNA probe/protein complexes was assessed by supershift analysis using antibodies against the p65 or p50 subunits of NF-κB or the unrelated transcription factor ATF2 (asterisks; lanes 9–11).

Fig. 3. ERK activation by Tpl2-transduced CD40 signals is MEK dependent. (A) A representative western blot of Tpl2^+/+ and Tpl2^–/– B-cell lysates harvested before and after anti-CD40 stimulation was probed with an antibody against the phosphorylated form of MEK (upper) or with an antibody that recognizes total MEK (lower). (B) In vitro kinase assays were carried out on anti-ERK immunoprecipitates derived from unstimulated and anti-CD40-stimulated Tpl2^+/+ and Tpl2^–/– B cells using GST–Elk1 as a substrate. Tpl2^+/+ B-cell cultures were also pretreated for 45 min with 20 µM MEK inhibitor PD98059 before being stimulated with the anti-CD40 antibody for 15 min.

Fig. 4. ERK activation by TNF-α in macrophages depends on signals transduced via Tpl2. (A) Representative western blots of Tpl2^+/+ and Tpl2^–/– bone marrow-derived macrophage (BMDM) cell lysates, harvested at the indicated time points before and after PMA or murine TNF-α (mTNFα) stimulation, was probed with antibodies to phosphorylated ERK (upper) or total ERK (lower). (B) Tpl2^+/+ and Tpl2^–/– BMDMs express both TNFR1 and TNFR2, as determined by RT–PCR. Amplification of the housekeeping gene HPRT serves as a cDNA synthesis and amplification control, and the lack of DNA contamination is confirmed by the absence of amplification in the RT– sample. (C) TNFR1 transduces ERK signaling in mTNF-α-treated mouse BMDMs. BMDMs from Tpl2^+/+ mice were pretreated for 30 min with 2 µg/ml blocking TNFR1, TNFR2 mAbs or a combination of these reagents and then stimulated for 15 min with 50 ng/ml mTNF-α. Lysates were examined for phosphorylated or total ERK. (D) The effects of human TNF-α (hTNFα), which exclusively stimulates the mTNFR1, on ERK activation in Tpl2^+/+ and Tpl2^–/– BMDMs.

Fig. 5. Tpl2 functions downstream of TRAF6 in the transduction of ERK activation signals. (A) CD40-mediated ERK activation is significantly impaired in mouse embryo fibroblasts (MEFs) isolated from Tpl2^–/– mice. Tpl2^+/+ and Tpl2^–/– cells were transiently transfected with CD40 or control vector in combination with HA-ERK1. The transfected cultures were serum starved and stimulated with recombinant soluble CD40L as indicated. Lysates were examined for expression of HA-ERK1 and phosphorylated ERK1 by immunoblotting. (B) Tpl2 is required for TRAF6-induced ERK activation. MEFs from Tpl2^+/+ and Tpl2^–/– mice were transiently transfected with HA-ERK1 together with FLAG-tagged expression vectors for TRAF2 or TRAF6. Lysates were examined for expression of FLAG-TRAF, HA-ERK1 and phosphorylated ERK1 by immunoblot. (C) Tpl2 is essential for TRAF6-induced ERK activation in mouse keratinocytes. SV40-transformed keratinocytes from Tpl2^+/+ and Tpl2^–/– mice were transfected with HA-ERK1 and increasing amounts of TRAF6. ERK activation was determined by anti-HA immune-complex in vitro kinase assay (IVK) using GST–Elk1 as a substrate. Data are representative of four independent experiments. (D) TRAF6-induced NF-κB transactivation, assessed by luciferasse reporter assays, is not impaired in Tpl2^–/– MEFs. Relative luciferase value (RLV) is the ratio of luciferase versus β-galactosidase measurements from duplicate determinations. Data are means ± SD from three independent experiments. (E) Endogenous Tpl2 and TRAF6 coimmunoprecipitate with aggregated CD40. Lysates from BJAB lymphoma cells stimulated with CD40L for 10 min or from untreated cultures were immunoprecipitated with the anti-CD40 mAb S2C6 (lanes 1 and 2) and probed with antibodies against TRAF6, Tpl2 or CD40. Lane 3 represents 40 µg of whole BJAB cell lysates (WCL) for Tpl2 and CD40 and 50 µg for TRAF6 detection. (F) CD40 augments the association of coexpressed Tpl2 and TRAF6. 293T cells were cotransfected with FLAG-tagged TRAF6, myc-tagged Tpl2 and/or CD40 as indicated. Cell lysates were immunoprecipitated with a FLAG antibody immobilized on G Sepharose beads. Immunoprecipitates were resolved in 9% SDS/PAGE and probed with either anti-myc (upper) or anti-TRAF6 antibody (lower).

Fig. 6. Cellular proliferation and expression of activation markers in Tpl2^+/+ and Tpl2^–/– B cells following CD40 stimulation. (A) Proliferation in unstimulated (NT) and anti-CD40-stimulated splenic B cells from three wild-type (WT) and three Tpl2^–/– (KO) mice was assessed by [³H]thymidine incorporation. Mean values (±SD) of triplicate determinations from a representative experiment are shown. Four independent experiments were performed and yielded similar results. (B) Induction of CD54, CD69, CD44 and CD95 in anti-CD40-stimulated Tpl2^+/+ and Tpl2^–/– B cells was examined by flow cytometry. Mean fluorescence intensity (mfi, mean values ± SD from three independent experiments) is shown.

Fig. 7. B cells from Tpl2^–/– mice are partially impaired in IgE production following CD40 and IL-4 stimulation. (A) Secreted IgE levels (means ± SD from seven independent experimental B-cell cultures) are reduced in Tpl2^–/– B cells stimulated with anti-CD40 plus IL-4. This difference is statistically significant (p = 0.016). (B) Secreted IgE levels (means ± SD from four independent experimental B-cell cultures) are reduced in Tpl2^+/+ B cells treated with IL-4 and anti-CD40 in the presence of the MEK inhibitor UO126. The levels of IgE from vehicle control (DMSO)-treated cultures are also shown. (C) CD40 does not affect the levels of STAT6 phosphorylation induced by IL-4. B cells from Tpl2^+/+ and Tpl2^–/– mice were stimulated with anti-CD40, IL-4 or anti-CD40 plus IL-4 for 15 min, and cell lysates were analyzed for the phosphorylation status of STAT6 by immunoblot analysis (upper). Equal loading was confirmed by probing the same blot for β-tubulin (lower). (D) IL-4 does not modify the MAPKs and NF-κB activation by CD40. B cells from Tpl2^+/+ and Tpl2^–/– mice were stimulated with anti-CD40, IL-4 or anti-CD40 plus IL-4 for 15 min, and cell lysates were analyzed for MAPK activation using antibodies that recognize either the phosphorylated or total forms of ERK, JNK and p38. IκBα degradation was assessed by immunoblot using an anti-IκBα polyclonal antibody. The experiments in (C) and (B) were carried out on the same cell lysates. This confirmed that both the anti-CD40 antibody and the IL-4 were functionally active when used alone.