|Home | About | Journals | Submit | Contact Us | Français|
Protein kinases are important regulators of intracellular signal transduction pathways and play critical roles in diverse cellular processes. TAK1, a member of the MAPKKK family, is essential for TNFα-induced NF-κB activation. Phosphorylation and Lys63-linked polyubiquitination (polyUb) of TAK1 are critical for its activation. However, whether TAK1 is regulated by polyubiquitination-mediated protein degradation after its activation remains unknown. Here we report that TNFα induces TAK1 Lys48 linked polyubiquitination and degradation at the later time course. Furthermore, we provide direct evidence that TAK1 is modified by Lys48-linked polyubiquitination at lysine-72 by mass spectrometry. A K72R point mutation on TAK1 abolishes TAK1 Lys48-linked polyubiquitination and enhances TAK1/TAB1 co-overexpression-induced NF-κB activation. As expected, TAK1 K72R mutation inhibits TNFα-induced Lys48-linked TAK1 polyubiquitination and degradation. TAK1 K72R mutant prolongs TNFα-induced NF-κB activation and enhances TNFα-induced IL-6 gene expression. Our findings demonstrate that TNFα induces Lys48-linked polyubiquitination of TAK1 at lysine-72 and this polyubiquitination-mediated TAK1 degradation plays a critical role in the downregulation of TNFα-induced NF-κB activation.
Protein kinases are important regulators of intracellular signal transduction pathways and play critical roles in a plethora of cellular processes such as cell cycle regulation, apoptosis and immune response [1-3]. The human genome encodes over 500 protein kinase genes which constitute approximately 2% of human genes . Perturbation of protein kinases can cause many types of human cancers . Once a kinase is activated, its activity will be subsequently downregulated through negative feedback systems to attenuate or terminate kinase activity and the signaling pathway that it regulates. Several mechanisms, including endocytosis, reduction of the kinase mRNA, dephoshorylation and ubiquitin-proteasome system, are involved in the negative regulation of kinase mediated signal transduction pathway [6-12].
Tumor necrosis factor α (TNFα) is a proinflammatory cytokine that plays a critical role in innate immune response by eliciting inflammation. Stimulation of cells with TNFα activates intracellular signaling pathways which lead to the activation of transcription factors such as NF-κB . Upon binding to its receptor, TNFα induces the formation of a receptor-associated complex composed by several adaptor proteins (TRADD, TRAF2, and TRAF5) and RIP1, which promotes Lys63-linked polyubiquitination and subsequent activation of RIP1 and TRAF2 [15-17]. The activated RIP1 and TRAF2 lead to TAK1 Lys63-linked polyubiquitination which is vital for its activation and signal transduction [18-22]. However, genetic evidence suggests that Tak1 but not Ripk1 is essential for TNFα-induced NF-κB activation in mouse embryonic fibroblasts (MEFs) [23-25]. Upon Lys63-linked polyubiquitination and activation, TAK1 causes activation of IκB kinases (IKKs) to activate transcriptional factor NF-κB .
Ubiquitin proteasome system plays a pivotal role in the regulation of NF-κB pathway [9, 27-29]. For instance, after RIP1 activation, A20 ubiquitin editing complex targets RIP1 for inactivation by promoting Lys48-linked polyubiquitination and degradation of RIP1 [30, 31]. Upon IKKβ activation, Ro52 catalyzes the monoubiquitination of IKKβ to down-regulate NF-κB signaling while at resting stage the IKKβ protein level is tightly regulated by another E3 ligase KEAP1 which promotes IKKβ proteolysis [32-34]. TAK1 is an essential signaling mediator that bridges TNFR1 to the IKKs complex. Whether TAK1 is also under the surveillance of ubiquitin proteasome system and more importantly, exactly how it is regulated by such a system remain to be elucidated. Here, we report that TNFα induces Lys48-linked polyubiquitination of TAK1 at lysine-72 and this polyubiquitination-mediated TAK1 degradation plays an important role in the downregulation of TNFα-induced NF-κB activation. Collectively, our data reveal a new regulatory mechanism for inactivation of TNFα-induced NF-κB activation.
293T cells and Tak1-deficient mouse embryonic fibroblast (MEF) cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 μg/ml streptomycin at 37 °C in 5% CO2. The following antibodies and reagents were used: TAK1 (4505), phospho-IKKα/β (2681), IKKβ (2684) phospho-JNK (9251), JNK (9252L), phospho-IκBα (9246L), IκBα (9242) and secondary antibodies conjugated to horseradish peroxidase (Cell Signaling); HA (sc-7392), RelA (sc-8008), Ubiquitin (sc-8017), PCNA (sc-56), and Protein A-agarose beads (Santa Cruz Biotechnology); FLAG-M2 (F3165) and β-actin (A2228) (Sigma); Lys63-specific Ubiquitin (05-1308 and 05-1313) and Lys48-specific Ubiquitin (05-1307) (Millipore); Recombinant mTNFα (R & D Systems); MG132 (Calbiochem).
Human TAK1 cDNA was inserted into the expression vector pcDNA3.1 with a Flag tag at its N-terminal. Site-directed mutagenesis was used to generate K72R point mutation. Retroviral TAK1 wild-type and K72R mutant constructs were generated using pBabe-puro vector. The NF-κB-dependent firefly luciferase reporter and pCMV promoter-dependent Renilla luciferase reporter plasmids were purchased from Clontech. HEK-293T cells and TAK1-deficient MEF cells were transfected with expression plasmids using FuGene 6 (Roche) and Lipofectamine 2000 (Invitrogen), respectively.
Topo-Vector, Topo-Flag-TAK1 wild-type and Topo-FLAG-TAK1 K72R mutant were transfected in to Tak1-deficient MEF cells. After 48 hours, transfected cells were selected by G418 (4 mg/ml) for 10 days. For retroviral infection, the pBabe empty vector, pBabe-TAK1 wild-type, pBabe-TAK1 K72R mutant were co-transfected with retrovirus packing vector Pegpam 3e and PLC-ECO in HEK-293T cells to obtain retroviral supernatants. Viral supernatants were collected after 48 and 72 hours. TAK1-deficient MEF cells were incubated with retroviral supernatant in the presence of 4 μg/ml polybrene. Stable cell lines were established after 10 days of puromycin (3 μg/ml) selection.
Cells were washed 3 times with ice-cold PBS, then lysated on ice with lysis buffer (25 mM HEPES (pH 7.7), 135 mM NaCl, 3 mM EDTA, 1% Triton X-100, 25 mM β-glycerophosphate, 0.1 mM sodium orthovanadate, 1 mM phenylmethylsulfonylfluoride, 1 mM dithiothreitol, 10 μg/ml aprotinin, 10 μg/ml leupeptin, 1 mM Benzamidine, 20 mM disodium p-nitrophenylphosphate, phosphatase inhibitor cocktail A and B from Sigma) for 30 min. After lysis, the whole protein extracts were collected by high-speed centrifuge at 15,000 g for 15 min at 4°C. For immunoprecipitation, primary antibodies were added to the supernatant and incubated with rotation for 3 h at 4°C. After brief centrifugation at 15,000 g for 5min, the supernanant was carefully transferred to be a new tube and incubated with protein A-agarose beads for 2 hours at 4°C. After binding, beads were washed 3 times with ice-cold lysis buffer and boiled in the laemmli buffer for 10min. The immunoprecipitates were resolved on a SDS-PAGE and transferred to nitrocellulose membranes for immunoblotting.
Five dishes (10cm in diameter) of 293T cells were transiently transfected with Flag-TAK1 expression vector. Cells were lysed in NETN (20 mM Tris-Cl (pH 8.0), 100 mM NaCl, 0.5 mM EDTA, 0.5 % (v/v) Nonidet P-40 (NP-40)) buffer containing 1% SDS (Sodium Docecyl Sulfate), and then boiled at 95 degree for 10 minutes to dissociate the interaction proteins. After boiling, the lysis was diluted 10 fold in NETN, centrifuged at 15,000 rpm and incubated with anti-FLAG antibody to immunoprecipitate Flag-TAK1.The immunoprecipitates were then eluted and boiled in SDS laemmli buffer, resolved on a SDS PAGE and gel regions above 70 kDa were cut and in-gel digested by trypsin as previously described [9, 27, 28]. Purified peptides were analyzed by a Linear Tandem Quadruple (LTQ) mass spectrometer (Thermo Fisher) equipped with a nanospray ESI source and an online HPLC system coupled to an auto sampler. The purified peptides were injected in a self-packed C18 column (5cm bedding,75 μm inner diameter and 5 μm particle size) and eluted with 55-min gradient of 95% solvent A (0.1% formic acid in water) to 30% solvent B (0.1% formic acid in acetonitrile) with a flow rate of ~450nl/min. A data-dependent experiment was carried out where the top 20 most intensive ions were triggered by collision induced disassociation (CID) for MS2. The precursor isolation window was set as 3 units. Raw data were searched for lysine ubiquitination against a human ref database (2009.06.27) by Sequest embedded in the Bioworks Browser (Thermo Fisher). Key parameters used for the search are: precursor and fragment mass tolerances were confined within 4.0 Da and 0.5 Da, respectively and a maximum of two missed cleavage sites were allowed. Other parameters for search include: a fixed cystein-destreak (+75 Da), and dynamic modifications on lysine ubiquitination (+114.1 Da) and methionine oxidation (+16 Da). MS2 spectra were filtered (Xcorr>2) for doubly charged peptides and Xcor>3 for triply charged peptides and peptide probability< 0.01) to tentatively identify TAK1 K72 ubiquitination and ubiquitin Lys48 linkages. These spectra were then manually inspected.
Total cellular RNA was isolated by TRIzol Reagent (Invitrogen) and 1 μg of RNA was used for each RT-PCR reaction. Quantitative real-time PCR was performed using specific primers and QTaq One-Step qRT-PCR SYBR kit (Clontech). Reactions were analyzed using an Applied Biosystems 7300 real time PCR system. Data were normalized to housekeeping GAPDH gene and the relative abundance of transcripts was calculated by the Ct models.
pBabe-vector, TAK1 wild-type and K72R mutant reconstituted TAK1-deficient MEF cells were either untreated or treated with mouse TNFα (2 ng/ml) for the indicated time points. Protein supernatants were collected after 15,000 g centrifuge for 15 min at 4°C. Mouse IL-6 concentrations in the supernatants were determined by ELISA (BD Biosciences) according to the manufacturer's instruction.
Cells were seeded at 3 × 105 cells per well and cultured overnight in 6-well plates. The cells were transfected with NF-κB-dependent firefly luciferase construct, Renilla luciferase and expression vectors. Luciferase activity was measured according to the manufacturer's protocol (Promega). The relative luciferase activity was calculated by dividing the firefly luciferase activity and the Renilla luciferase activity. All the data represent three independently repeated biological experiments performed in triplicate.
We previously found that in response to TNFα, TAK1 is quickly activated by Lys63-linked polyubiquitination. Interestingly, we also observed that TAK1 is modified by Lys48-linked polyubiquitination at the later stage (more than 30 mins after TNFα stimulation) of its activation . This observation prompts us to further investigate Lys48-linked polyubiquitination of TAK1. We stably re-introduced Flag-TAK1 wild-type into TAK1-deficient MEF cells and treated these cells with TNFα at different time points over a course of 2 hours in the presence of MG132. As expected, we found that TNFα induced TAK1 polyubiquitination in the presence of MG132, while without TNFα stimulation, MG132 alone was not sufficient to induce TAK1 polyubiquitination (Fig. 1a). TAK1 polyubiquitination induced by TNFα and MG132 is Lys48-linked because it is specifically recognized by Lys48- but not Lys63- specific antibodies (Fig. 1a). The time-dependent accumulation of TAK1 Lys48-linked polyubiquitination indicates that it mediates TAK1 degradation. Next we examined whether TNFα could induce TAK1 degradation in HeLa cells. We incubated the cells with cycloheximide (CHX) for 2 hours to block protein synthesis. Consistent with the observation that TNFα could induce TAK1 polyubiquitination, TNFα promoted degradation of a portion of TAK1 protein in one hour of treatment (Fig. 1b). In addition, we also found that TNFα could slightly induce IKKβ degradation and potently trigger the degradation of TNFR1 as well as IκBα upon 1-hour treatment of TNFα. (Fig. 1b-c). Together, our results indicate that TNFα induces TAK1 polyubiquitination and degradation at the later time course of stimulation.
To obtain the direct evidence that TAK1 is ubiquitinated by Lys48-linked polyubiquitination and to map the precise lysine ubiquitination site(s), we immunoprecipitated the Flag-TAK1 under partial protein denaturing conditions (see materials and methods for details), separated the immunoprecipitates by SDS PAGE, digested in gel by trypsin and employed mass spectrometry for detection of lysine ubiquitination site(s). To accumulate primarily the Lys48-linked polyubiquitinated TAK1 conjugates, HEK-293T cells were treated by MG132 for 2 hours before harvesting (Fig. 2a). At higher molecular weight region where polyubiquitinated TAK1 was supposed to be accumulated, we were able to identify multiple peptides of Lys48 poly-Ub linkages (Fig. 2b) as well as an ubiquitianted TAK1 peptide that is conjugated at lysine 72 by diGlycine isopeptide bond (ubiquitination signature) using mass spectrometry (Fig. 2c). We were unable to detect any other polyUb linkage in the TAK1 immunoprecipitates under such condition, indicating that K48 polyUb is likely the dominant form of polyUb chain that targets primarily at K72 position. Based on the crystal structure , TAK1 lysine-72 localizes on the surface of the protein where the ubiquitination conjugation likely occur (Fig. S1). These results suggest that TAK1 is modified by Lys48-linked polyubiquitination at lysine-72 which may play a role in modulating TAK1 activity.
To confirm that TAK1 is modified by Lys48-linked polyubiquitination, we co-transfected FLAG-TAK1 into HEK-293T cells with HA-ubiquitin wild-type, Lys48 or Lys63 only mutants. Transfected cells were treated with/without MG132 for proteasome inhibition. Flag-TAK1 was then immunoprecipitated from different transfected cell lines and blotted with anti-HA, anti-Lys48 and anti-Lys63 specific ubiquitin antibodies for the detection of ubiquitination smears. Consistent with MS data, we found that TAK1 was modified by Lys48-linked polyubiquitination and the ubiquitination TAK1 was accumulated upon proteasome inhibition (Fig. 3a).
To ascertain that lysine-72 is the major ubiquitination conjugation site for TAK1 Lys48-linked polyubiquitination, we made a TAK1 mutant in which the lysine-72 is mutated to arginine (K72R) by site directed mutagenesis. Flag-TAK1 wild-type or K72R mutant along with HA-ubiquitin were transiently transfect into HEK-293T cells. Ubiquitinated TAK1 was then immunoprecipitated by anti-Flag antibodies and detected by anti-HA antibodies. As shown in Fig. 3b, this point mutation virtually completely abolished MG132-dependent TAK1 polyubiquitination. Together with the MS result, our data suggest that lysine-72 is the major lysine ubiquitination site for Lys48-linked TAK1 polyubiquitination.
To test if lysine-72 is indeed a functional ubiquitination site, we transiently transfect TAK1 K72R mutant and TAB1 as well as NF-κB luciferase reporter constructs into 293T cells. As shown in Fig. 3c, overexpression of K72R mutant induced a higher NF-κB-dependent luciferase reporter activity compared to that of TAK1 wild-type. Together, these results indicate that Lys48-linked TAK polyubiquitination at lysine-72 negatively regulate TAK1 function.
To test the hypothesis that lysine-72 polyubiquitination of TAK1 negatively regulates TAK1 activation in the later time point of TNFα stimulation, we stably re-introduced Flag-TAK1 wild-type and K72R mutant into TAK1-deficient MEF cells and treated these two cells with TNFα at four consecutive time points (at 30, 60, 90 and 120 mins) in the presence of MG132. As shown in Fig. 4a, polyubiquitination of wild type TAK1 were potently induced after 30 minutes of TNFα stimulation . However, similar scenario did not replicate in the K72R mutant cells, in which no detectable TAK1 polyubiquitination induction and accumulation were observed.
To test whether this polyubiquitination at lysine-72 represents the signal for TAK1 protein degradation, we treated either TAK1 wild-type- or K72R mutant-reconstituted MEF cells with TNFα at different time points in the presence of CHX to inhibit TAK1 protein biosynthesis. Consistently, we found that TNFα-induced TAK1 degradation was inhibited in TAK1 K72R mutant cells comparing to that of wild type cells. The protein level of wild type TAK1 but not K72R was appreciably reduced in 6 hours of TNFα stimulation (Fig. 4b-c). These results suggest that TAK1 Lysine-72 residue mediates TNFα-induced TAK1 polyubiquitination and degradation at the later stage of TNFα stimulation.
To further characterize the functional role of TAK lysine-72-mediated polyubiquitination in the context of TNFα-mediated NF-κB pathway, we probed the TNFα-induced IKKs phosphorylation, JNK phosphorylation, IκBα phosphorylation, IκBα degradation, and RelA nuclear translocation as markers for the activation of the pathway. TNFα-induced phosphorylation levels of IKKs, JNK and IκB as well as RelA nuclear presence are considerably higher and last longer in TAK1 K72R mutant-reconstituted cells compared to the wild-type counterpart (Fig. 5a-b). Consistent with Figure 3c, we also found that TAK1 K72R point mutation enhanced TNFα-induced NF-κB luciferase activity compared to that of TAK1 wild-type (Fig. 5c). Taken together, these results indicate that TAK1 polyubiquitination at lysine-72 is functional and might be important for preventing excessive TNFα-induced NF-κB activation, which could be detrimental to the cells.
TNFα induces IL-6 gene expression via NF-κB activation. To further examine the role of TAK1 polyubiquitination at lysine-72 in TNFα-induced NF-κB targeted gene expression, total RNAs were isolated from either TAK1 wild-type or K72R mutant stable MEF cell lines with or without TNFα stimulation and analyzed by quantitative RT-PCR. In this assay, we found that TNFα induced a higher level of IL-6 gene expression and IL-6 protein production in the TAK1 K72R mutant-reconstituted cells compared to the TAK1 wild-type-reconstituted cells (Fig. 6a-b). These datasuggest that polyubiquitination of TAK1 at lysine-72 is required for preventing excessive TNFα-induced NF-κB activation and cellular responses.
Deregulation of immune responses can be extremely harmful to the host cells. As a master transcription factor in immunity, NF-κB activity must be tightly controlled. TAK1, a member of the MAPKKK family, was originally found to function in the transforming growth factor-β (TGF-β)-mediated MAPK activation . TAK1 has been further demonstrated to be an essential component in TNFα-mediated activation of NF-κB and JNK [37-39]; Post-translational regulation of TAK1, especially via phosphorylation, has been extensively investigated. For example, it has been suggested that after it own activation, TAK1 is dephosphorylated and subsequently inactivated by phosphotases such as PP2A and PP6 [40, 41]. However, whether TAK1 is modulated by the ubiquitin/proteasome pathway is unclear. In this report, we found that TAK1 is modified by Lys48-linked polyubiquitination and degraded by ubiqutin proteasome system to downregulate TAK1-mediated NF-κB activation. Furthermore, we provided evidence that TAK1 lysine 72 is the major site for TAK1 Lys48-linked polyubiquitination. Lastly, we presented data to show that this modification is functional and important for the inhibition of the pathway. Thus, our data demonstrate that ubiquitin/proteasome pathway-mediated TAK1 degradation plays a critical role in the downregulation of TNFα-induced NF-κB activation.
Innate immunity provides an evolutionarily conserved barrier for higher eukaryotes to battle against infectious microorganisms. Genetic approaches in Drosophila have uncovered two distinct pathways that are mediated by Toll and Imd. TAK1 acts as an activator of both the JNK and the Relish in Imd signaling. It has been suggested that proteasomal degradation of TAK1 by an E3 ligase POSH could lead to rapid termination of JNK activation in Drosophila cells . Two other ubiquitin E3 ligases have been proposed to target TAK1 for ubiquitin-proteasome degradation in mammalian system, although the detailed molecular mechanism for such regulation is still murky . RNF142/SH3RF1 is a human homolog of fly POSH protein. Whether or not RNF142 is the primary E3 ligase for TAK1 K72 ubiquitination shall be tested in the future
Ubiquitination via different linkages may play distinct roles [29, 44]. While Lys48-linked polyubiquitination is known for substrate degradation, Lys63-linked polyubiquitination regulates non-proteolysis functions of the substrates such as control of kinase activity in the signaling transduction pathway . For instance, recruitment of endogenous RIP1 to TNFR1 is initiated by the transient Lys63-linked polyubiquitination of RIP1 which is then followed by Lys48-linked polyubiquitination to terminate the signal tansduction pathway . While we could not see obvious TNFα-induced RIP1 degradation after 90 minutes of TNFα treatment, we did observe a minor yet truly detectable amount of TNFα-induced degradation of TAK1 and IKKβ. It is likely that the small portion of degradable TAK1 proteins are the activated TAK1 during signal transduction and degradation of the activated TAK1 is critical for the termination of TAK1-mediated signaling. Indeed, similar phenomenon of “ubiquitin editing” was observed by in IL-1β-induced IRAK1 polyubiquitination . In TNFα-mediated signaling, Lys63-linked polyubiquitination of TAK1 is required for its activation while Lys48-linked polyubiquitination of TAK1 leads to its degration. It is highly likely that such ubiquitin editing mechanism is common for key molecules in TNFα-induced NF-κB activation.
In summary, we present evidence to suggest that Lys48-linked TAK1 polyubiquitination at lysine-72 negatively regulates TNFα-induced NF-κB activation. In view of the data presented here and in previous reports, we propose a working model (Fig. 7) in which, upon TNFα binding to its receptor, TNFα receptor-mediated signaling events lead to Lys63-linked TAK1 polyubiquitination at lysine-158 and its immediate activation. After activation, TAK1 is further modified by Lys48-linked polyubiquitination at lysine-72 to downregulate TAK1-mediated NF-κB activation.
This work was supported, in whole or in part, by the NIH/NINDS grant 1R01NS072420-01 (to J.Y.), the Virginia & L E Simmons Family Foundation Collaborative Research Fund (to J.Y.), Dan L. Duncan Cancer Center Pilot Project (to J.Y.), and the NIH CA84199 (to J.Q.) and Welch Foundation (to J.Q.). The National Basic Reseach Program (973 Program, No. 2007CB48000; No. 2009CB941100; and No. 2011CB965102) (to G.X.)