Jun Turnover Is Controlled Through JNK-Dependent Phosphorylation of the E3 Ligase Itch

See allHide authors and affiliations

Science  08 Oct 2004:
Vol. 306, Issue 5694, pp. 271-275
DOI: 10.1126/science.1099414


The turnover of Jun proteins, like that of other transcription factors, is regulated through ubiquitin-dependent proteolysis. Usually, such processes are regulated by extracellular stimuli through phosphorylation of the target protein, which allows recognition by F box–containing E3 ubiquitin ligases. In the case of c-Jun and JunB, we found that extracellular stimuli also modulate protein turnover by regulating the activity of an E3 ligase by means of its phosphorylation. Activation of the Jun amino-terminal kinase (JNK) mitogen-activated protein kinase cascade after T cell stimulation accelerated degradation of c-Jun and JunB through phosphorylation-dependent activation of the E3 ligase Itch. This pathway modulates cytokine production by effector T cells.

Ubiquitin-dependent proteolysis controls turnover and abundance of transcription factors and other regulatory proteins (1). Protein ubiquitination requires the concerted action of ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzymes (E2), and ubiquitin ligases (E3) (2, 3). Extracellular stimuli can regulate protein turnover through inducible substrate phosphorylation which confers recognition by F box–containing E3 ligases (4). Such E3 ligases, which are devoid of catalytic activity, recognize only the phosphorylated forms of their substrates (5). Transcription factors regulated through ubiquitin-dependent turnover include the Jun proteins, components of the AP-1 transcription factor. The activity of c-Jun and JunB is enhanced by phosphorylation of their transcriptional activation domain by JNKs (6, 7). JNK-dependent phosphorylation can also stabilize c-Jun (8, 9). Recently, however, JNK-mediated phosphorylation was shown to accelerate c-Jun degradation by allowing its recognition by the E3 ligase Fbw7-containing Skp/Cullin/F-box protein complex (SCFFbw7) (10). Here, we provide physiological and biochemical evidence for another pathway through which extracellular stimuli control c-Jun and JunB abundance. This process is based on inducible phosphorylation of an E3 ligase of the homology to the E6-associated protein C terminus (HECT) family, which increases its catalytic activity.

The HECT domain E3 Itch functions in ubiquitin-dependent degradation of both c-Jun and JunB, which accumulate in T cells of Itchy mice, lacking Itch activity (11). Accumulation of JunB, a transcription factor that promotes differentiation of T helper 2 (Th2) cells and interleukin-4 (IL-4) gene transcription (12, 13), is likely to contribute to excessive Th2 cytokine production in Itchy mice (11). A similar phenotype is exhibited by T cells lacking JNK1 (14) or both JNK1 and JNK2 (15), suggesting that JNK activation negatively regulates Th2 cytokine production. It is not known, however, whether this function is related to JNK-dependent modulation of JunB turnover.

To understand how JNK activation regulates Th2 cytokine expression, we examined T cells from Mekk1ΔKD mice that express an inactive form of mitogen and extracellular kinase kinase 1 (MEKK1) (16), a mitogen-activated protein kinase (MAPK) kinase kinase (MAP3K) that is a potent activator of JNK signaling (17). Mekk1ΔKD mice are viable (16) without any obvious defect in generation or survival of T cells or their intrathymic differentiation into CD4+ and CD8+ subsets (fig. S1). Yet, T cells from these mice exhibit reduced JNK activation following engagement of the T cell receptor (TCR) and the CD28 auxiliary receptor (Fig. 1A). Mekk1ΔKD peripheral T cells and thymocytes also hyperproliferated in response to stimulation with antibodies to CD3 and CD28 (fig. S2) and within 4 hours of receptor engagement expressed larger amounts of IL-4 and IL-13 mRNAs relative to those in wild-type (WT) cells, although they expressed normal amounts of IL-2 mRNA (Fig. 1B). IL-4, as the master regulator of Th2 effector T cell differentiation, can induce expression of other cytokines, including IL-5, IL-10, and IL-13 (18). Indeed, after 24 hours, activated Mekk1ΔKD T cells expressed increased amounts of IL-5, IL-10, and IL-13 mRNAs in addition to IL-4 mRNA (Fig. 1B). The mRNAs for IL-2 and the Th1 cytokine interferon- γ (IFN γ) remained unchanged. We also cultured naïve WT or Mekk1ΔKD CD4+ T cells under Th1- or Th2-polarization conditions (18, 19) to examine their differentiation into effector cells. Because of the BL6×C129 genetic background (20), WT Th2 cells produced relatively small amounts of IL-4, but their amounts were increased in Mekk1ΔKD Th2 cells (Fig. 1C). Production of IFN γ by Th1 cells remained unchanged.

Fig. 1.

Increased Th2 cytokine production in Mekk1ΔKD T cells with reduced JNK activity. (A) WT and Mekk1ΔKD T cells were incubated with antibodies to CD3 and CD28. When indicated, JNK activity was measured by an immunecomplex kinase assay with GST–c-Jun as the substrate. Phosphorylated c-Jun was detected by autoradiography and quantitated with a PhosphorImager. The amount of immunoprecipitated JNKs was determined by immunoblotting. Relative JNK activity (RA) at t = 0 was given an arbitrary value of 1.0. (B) CD4+ T cells from WT and Mekk1ΔKD mice were incubated with antibodies to CD3 and CD28. Cytokine mRNA levels were quantitated by real-time polymerase chain reaction at the indicated times and normalized to amounts of cyclophilin A mRNA. The relative amount (RA) of IL-4 mRNA in WT cells at each time point was given an arbitrary value of 1.0. (C) CD4+ T cells were cultured under Th1- or Th2-polarizing conditions for 7 days. Th subsets were restimulated with antibodies to CD3 and CD28 and analyzed for cytokine expression by flow cytometry after 6 hours. (D) CD4+ T cells from Jnk1–/– and Mekk1+/ΔKDJnk1+/– mice were stimulated with antibodies to CD3 and CD28 and cytokine mRNAs were quantitated as in (B) after 24 hours. Error bars in (B) and (D) show mean + SD.

To examine whether the effect of MEKK1 on Th2 cytokine production is JNK dependent, we crossed Jnk1–/– mice with Mekk1ΔKD mice. Mekk1+/ΔKDJnk1+/– CD4+ T cells expressed increased amounts of IL-4, IL-5, IL-10, and IL-13 mRNAs relative to those in Jnk1+/– (Fig. 1D) or Mekk1+/ΔKD (21) T cells, indicating JNK1 dependence. These results agree with Dong et al., who found hyperproliferation and increased Th2 cytokine production by Jnk1–/– T cells (14), and differ from those reported by Sabapathy et al., who found reduced proliferation of Jnk1–/– T cells, attributable to diminished IL-2 expression (22). As shown above, the loss of MEKK1 and JNK1 does not affect IL-2 mRNA levels. The elevated Th2 cytokine response of Mekk1ΔKD CD4+ T cells was IL-4 dependent, given that it was absent in Mekk1ΔKDIL-4–/– T cells (fig. S3).

Multiple transcription factors, including nuclear factor of activated T cells (NF-AT), GATA-3, c-Maf, and JunB, regulate IL-4 gene expression (19). We examined the amounts of these transcription factors in WT and Mekk1ΔKD T cells. Both c-Jun and JunB were increased in activated Mekk1ΔKD CD4+ T cells, whereas amounts of JunD, another family member, as well as GATA-3, c-Maf, NF-ATc1, and NF-ATc2 were unchanged (Fig. 2A). The effect on c-Jun and JunB resulted from loss of JNK activity, because two different JNK inhibitors—SP600125 (23) and JNKI-1, a peptide inhibitor based on the c-Jun docking site for JNK (24, 25)—also increased c-Jun and JunB expression (Fig. 2B). No such effects were observed with the p38 inhibitor SB202190 (21). Elevated c-Jun and JunB expression was also found in Jnk1–/– T cells, whereas Jnk2–/– T cells showed a slight reduction (fig. S4). These results indicate that a JNK1 deficiency rather than a JNK2 deficiency is responsible for the altered expression of c-Jun and JunB in Mekk1ΔKD T cells.

Fig. 2.

Decreased turnover of JunB and c-Jun after inactivation of MEKK1 or JNK. (A) WT and Mekk1ΔKD CD4+ T cells were stimulated with antibodies to CD3 and CD28 for 24 hours. Cell extracts were prepared and the amounts of various transcription factors were measured by immunoblotting and densitometry. The amount of each protein in WT cells was given an arbitrary value of 1.0. (B) WT CD4+ T cells were stimulated as above in the absence or presence of JNK inhibitors (SP600125 or JNKI-1) for 24 hours. Jun protein levels were analyzed as above and JNK activity was measured by an immunecomplex kinase assay. (C) The mRNA amounts of Jun family members and c-Fos in cells from (A) were measured by ribonuclease protection. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. (D) WT and Mekk1ΔKD T cells were pulse-labeled with [35S] amino acids and chased with nonlabeled amino acids. Proteins were immunoprecipitated at the indicated times postlabeling, separated by SDS–polyacrylamide gel electrophoresis and analyzed by autoradiography and phosphorimaging. The relative amounts of each [35S]-labeled protein at t = 0 were considered to be 100%. (E) WT CD4+ T cells were stimulated with antibodies to CD3 and CD28. At the indicated time points, the relative levels of JunB mRNA and protein were quantitated as in (A) and (C). Error bars in (E) and (F) show mean + SD.

Despite the change in protein abundance, the amounts of c-Jun and JunB mRNAs remained unchanged (Fig. 2C). We therefore examined whether MEKK1 and JNK promoted c-Jun and JunB turnover. In pulse-chase experiments, in WT T cells, newly synthesized c-Jun and JunB turned over with half-lives (T1/2) of 59 and 78 min, respectively, but were more stable in Mekk1ΔKD cells, with T1/2s of 109 and 280 min, respectively (Fig. 2D). The turnover of RelA(p65) remained unaltered. After engagement of TCR and CD28 in WT and Mekk1ΔKD T cells, the amounts of JunB mRNA were very similar, but the amounts of JunB protein after 1 to 4 days of stimulation were greater in the mutant cells (Fig. 2E). Thus, activation of the JNK MAPK cascade appears to accelerate turnover of JunB in CD4+ T cells and may have an instructional role in their polarization into Th1 and Th2 effector cells.

The phenotypic and biochemical similarities between Mekk1ΔKD and Itchy T cells raised the possibility that JNK-enhanced JunB and c-Jun turnover is Itch dependent. We examined this in transfected 293T cells because of difficulties in detecting endogenously ubiquitinated proteins in primary T cells. Ectopic expression of MEKK1 in 293T cells promoted polyubiquitination of both c-Jun (21) and JunB (Fig. 3A) in an Itch-dependent manner. Overexpression of kinase-deleted MEKK1 did not enhance Itch-dependent Jun polyubiquitination (Fig. 3A). Itch-dependent Jun ubiquitination was enhanced by a constitutively active JNKK2-JNK1 fusion protein, but not by an inactive version (Fig. 3B). JNK-enhanced ubiquitination correlated with accelerated JunB degradation (fig. S5). Treatment with a JNK inhibitor reduced the extent of Itch + MEKK1-induced polyubiquitination of c-Jun and JunB (Fig. 3C). This effect of JNK, however, appeared to be independent of JNK-mediated c-Jun phosphorylation because mutant versions of c-Jun lacking its JNK phosphorylation sites (8, 26) were ubiquitinated as efficiently as WT c-Jun in cells overexpressing Itch and WT MEKK1 (Fig. 3D).

Fig. 3.

The MEKK1-JNK cascade promotes c-Jun and JunB ubiquitination by enhancing Itch activity. (A) 293T cells were transiently transfected with plasmids encoding hemagglutinin (HA)–tagged ubiquitin, Myc-tagged JunB, WT or a kinase domain–deleted (mt) MEKK1, and WT or catalytically inactive (mt) Itch. Ubiquitin conjugation was examined by immunoblotting with an antibody to HA and quantitated by densitometry. IgG, immunoglobulin G. (B) 293T cells were transfected with HA-ubiquitin, Myc-tagged c-Jun or JunB, WT or catalytically inactive (mt) Itch, and WT or inactive (mt) JNKK2-JNK1 fusion proteins. Jun ubiquitination was examined as above. (C) 293T cells were transfected with c-Jun or JunB, HA-ubiquitin, Itch, and MEKK1. Cells were incubated with or without JNK inhibitor (SP600125) for 24 hours and analyzed as above. (D) 293T cells were transfected as above with WT or phosphorylation-deficient c-Jun constructs, Itch, and MEKK1. Ubiquitin conjugation was analyzed as above and c-Jun phosphorylation was examined by immunoblotting. (E) WT and Mekk1ΔKD T cells were left unstimulated or stimulated with antibodies to CD3 and CD28 for 15 min. Itch was immunoprecipitated and incubated with ubiquitin, E1, E2, and adenosine 5′-triphosphate (ATP). Itch self-ubiquitination was analyzed by immunoblotting with antibody to ubiquitin and quantitated by densitometry. (F) T cells were treated as above. Itch immunecomplexes were isolated and incubated with ubiquitin, E1, E2, ATP, and purified GST–c-Jun. c-Jun ubiquitination was analyzed by immunoblotting. (G) WT T cells were stimulated as above in the absence or presence of SP600125. In vitro ubiquitination assays with the use of immunoprecipitated Itch as the E3 with or without GST–c-Jun as the substrate were done as above.

Unlike F box–containing E3 ligases, HECT domain ligases are thought to recognize their substrates independently of their phosphorylation. We therefore examined whether JNK activation enhances Jun ubiquitination by modulating the activity of Itch. Itch undergoes self-ubiquitination, and this activity was enhanced if it was isolated from WT T cells activated with antibodies to CD3 and CD28 (Fig. 3E). Little enhancement of Itch self-ubiquitination was observed after activation of Mekk1ΔKD T cells (Fig. 3E). The ability of Itch to promote ubiquitination of a glutathione S-transferase (GST)–c-Jun substrate was enhanced in response to T cell activation and this response was also diminished in Mekk1ΔKD cells (Fig. 3F). Both Itch self-ubiquitination and its ability to promote c-Jun polyubiquitination largely depended on incubation with both E1 and E2 (Ubc7) enzymes (fig. S6) and were reduced in activated T cells that were treated with a JNK inhibitor (Fig. 3G).

To examine whether Itch is a target for JNK-mediated phosphorylation, we separated proteins from nonactivated and activated T cells by two dimensional (2D) gel electrophoresis and transferred them to membranes. Following T cell activation, Itch became more negatively charged and displayed a lower isoelectric point (Fig. 4A). These changes are consistent with increased Itch phosphorylation and were reversed by calf intestine alkaline phosphatase (CIAP) (Fig. 4B). The kinetics of Itch phosphorylation correlated with those of JNK activation (fig. S7). Furthermore, when Itch phosphorylation was reduced by treatment with a JNK inhibitor and compared with that of WT cells, less TCR-induced Itch phosphorylation was observed in Mekk1ΔKD cells (Fig. 4A). Similarly, Itch self-ubiquitination and c-Jun polyubiquitination were reduced after CIAP treatment of isolated Itch (Fig. 4C) (21). Moreover, incubation of Itch from non-stimulated T cells with activated JNK1 enhanced its self-ubiquitination, but incubation with inactive JNK1 had no effect (Fig. 4D). Incubation of Itch with active JNK1 also led to its efficient phosphorylation (Fig. 4E). Incubation of in vitro translated Itch with JNK1 also increased its self-ubiquitination and its ability to promote c-Jun polyubiquitination (Fig. 4F). Similar results were obtained with recombinant Itch produced in Escherichia coli (Fig. 4G). Consistent with the changes in c-Jun and JunB expression seen in Jnk1–/– and Jnk2–/– T cells (fig. S4), JNK1 is a more efficient Itch kinase than JNK2 and as a result is a more potent activator of Itch (fig. S8). The highly efficient phosphorylation of Itch by JNK1 is due to the presence of a JNK docking site, whose mutational inactivation prevents Itch phosphorylation and activation by JNK1 (21).

Fig. 4.

Increased E3 activity of Itch after JNK-mediated phosphorylation. (A) Proteins from non-stimulated (0') and activated (10') WT T cells incubated without or with SP600125 were resolved by 2D gel electrophoresis, transferred to membranes, and immunoblotted with antibodies to Itch and actin. Itch migration in activated Mekk1ΔKD cells was similarly analyzed. (B) Extracts from activated WT T cells were incubated without or with CIAP before 2D gel electrophoresis and immunoblotting as above. (C) Itch was immunoprecipitated from the extracts used in (B) and incubated with ubiquitin, E1, E2, and ATP. Itch ubiquitination was analyzed by immunoblotting with the antibody to ubiquitin and densitometry. (D) Immunoprecipitated Itch from nonstimulated T cells was incubated with active (wt) or inactive (mt) JNK1 and JNKK1. Itch ubiquitination was analyzed as above and quantitated by densitometry. (E) Immunoprecipitated Itch from nonstimulated T cells was incubated with active or inactive JNK1 and JNKK1 in the presence of [γ-32P]ATP. GST–c-Jun was included as a positive control. Protein phosphorylation was analyzed by autoradiography. The same membrane was probed with antibody to Itch. In vitro translated Itch (F) or a recombinant GST-Itch (G) was incubated with JNK1 and JNKK1. Itch and c-Jun polyubiquitination were analyzed as above. Itch phosphorylation was analyzed by autoradiography.

Extracellular stimuli often affect ubiquitin-dependent proteolysis by inducible target protein phosphorylation, which confers recognition by F box–containing E3 ligases (5). Instead, the effect of JNK on Itch-dependent c-Jun and JunB polyubiquitination and turnover is exerted by means of Itch itself. Some members of the HECT domain family are thought to be constitutively active E3 ligases. However, Itch catalytic activity is strongly modulated in response to T cell activation, through JNK-dependent phosphorylation. It is plausible that other members of the HECT domain family may be subject to similar regulation. The regulation is exerted at the level of the enzyme and not the substrate, and as is the case for F box-containing ligases, it allows a simultaneous increase in the turnover of multiple proteins. Disruption of Itch-dependent JunB turnover either through inactivation of Itch or inhibition of JNK results in increased expression of Th2 cytokine genes. Given the importance of JunB for IL-4 gene transcription and Th2 differentiation (12, 13), one function of JNK-dependent Itch activation is likely to be attenuation of IL-4 production in response to strong T cell activating signals.

Supporting Online Material

Materials and Methods

Figs. S1 to S8


References and Notes

View Abstract

Navigate This Article