TAK-ling IKK Activation: “Ub” the Judge

See allHide authors and affiliations

Science Signaling  19 Jan 2010:
Vol. 3, Issue 105, pp. pe3
DOI: 10.1126/scisignal.3105pe3


Investigation of the signaling events that lead to the activation of the transcription factor nuclear factor κB (NF-κB) has been a hotbed for the discovery of previously uncharacterized signaling mechanisms. The important role that nondegradative polyubiquitin chains play in these processes is now well recognized; however, precisely how they orchestrate NF-κB signaling is still a matter of much controversy. A recent study has challenged the dogmatic view by demonstrating that interleukin-1β (IL-1β), a major proinflammatory cytokine, activates two consecutive pathways, the “RING” and “zinc” pathways, to coordinate early and late activation of NF-κB, respectively. This study introduces a paradigm shift in the still-evolving mechanism of regulation of NF-κB.

Nuclear factor κB (NF-κB) is a central transcriptional regulator of inflammatory and immune responses, processes that are crucial for fighting off infections, repairing damaged tissues, and oncogenesis. Precise control of the activity of NF-κB is critical, because too little activity leads to insufficient protection against pathogens, whereas too much activity can contribute to autoimmune or inflammatory pathologies (1, 2). Numerous factors activate NF-κB, including the major proinflammatory cytokine interleukin-1β (IL-1β). Although activation of NF-κB may be very transient, long-term exposure to IL-1β causes both rapid (within minutes) and sustained (hours) activation of NF-κB. Whereas the mechanisms that control the initial acute phase of NF-κB activation have been intensively investigated, those that control sustained activation are not well understood. The study by Yamazaki et al. (3) provides a fresh view on the rapid activation of NF-κB and expands our understanding of the temporal regulation of IL-1β–dependent activation of NF-κB.

Activation of NF-κB by IL-1β involves engagement of the IL-1 receptor (IL-1R) and the IL-1R accessory protein as well as the recruitment of adaptor proteins, which ultimately results in the formation of an intermediate signaling complex between IL-1R–associated kinase 1 (IRAK1) and the E3 ubiquitin ligase tumor necrosis factor (TNF) receptor–associated factor 6 (TRAF6). TRAF6 leads to the activation of TAK1 [transforming growth factor–β (TGF-β)–activated kinase 1], which in turn phosphorylates and activates the inhibitor of NF-κB (IκB) kinase (IKK) to finally induce the activation of NF-κB (Fig. 1). TRAF6 is a multidomain protein that includes a really interesting new gene (RING) domain and several zinc finger domains (4). The RING domain has ubiquitin ligase activity that assembles a polyubiquitin chain in which successive ubiquitin molecules are linked through Lys63 (K63) to the previous ubiquitin molecule in the presence of the E2 ubiquitin-conjugating enzyme UBC13/UEV1A (5). These K63-linked polyubiquitin chains serve as a unique scaffold to recruit TAK1 through its ubiquitin-binding subunits TAK1-binding protein 2 (TAB2) or TAB3 and IKK through its ubiquitin-binding subunit IKKγ [also known as NF-κB essential modulator (NEMO)]. This coordinated recruitment of IKK and its upstream kinase TAK1 leads to the phosphorylation of IKK and the subsequent activation of NF-κB.

Fig. 1

Two consecutive NF-κB activation pathways induced by IL-1β. In the early RING pathway, engagement of IL-1R and IL-1R–associated protein by IL-1β results in the recruitment of myeloid differentiation primary response gene 88 (MyD88), IRAK4, IRAK1, and TRAF6 to the receptor. After phosphorylation of IRAK1 by IRAK4, an IRAK1-TRAF6 signaling complex may dissociate from the receptor complex. TRAF6 contains RING, zinc finger, coiled coil, and TRAF-C domains. TRAF6 is an E3 ubiquitin ligase and, together with UBC13/UEV1A (an E2 ubiquitin-conjugating enzyme), it forms K63-linked polyubiquitin chains (shown as repeating units in orange). Numerous proteins become K63-polyubiquitinated, including TRAF6, IRAK1, NEMO (not shown), TAB2/3 (not shown), and TAK1. K63-linked polyubiquitination of TAK1 is required for formation of the TRAF6-MEKK3-TAK1 complex. MEKK3 phosphorylates TAK1, which in turn phosphorylates IKKβ to induce phosphorylation of IκBα to cause its degradation by the ubiquitin-dependent proteasomal pathway to release active NF-κB (shown as a p50-p65 heterodimer). In the later zinc pathway, MEKK3 together with TRAF6 somehow causes the activation of IKK to induce degradation of IκBα and activation of NF-κB. These pathways are molecularly separable; however, it is possible that a DUB might disassemble K63-linked polyubiquitin chains, thereby linking the two pathways.

Several signaling proteins become modified by K63-linked polyubiquitination, including TRAF6 and IRAK1 (68). Yamazaki et al. showed that TAK1 was also rapidly K63-polyubiquitinated after stimulation of cells with IL-1β. Because the current model suggests that the role of K63-linked polyubiquitination in the activation of TAK1 is through TAB2-mediated recruitment, no obvious role for the modification of TAK1 by polyubiquitin was apparent. Indeed, when the authors looked for the function of this modification of TAK1, they were rewarded with exciting new discoveries. They first mapped the TAK1 modification site as Lys209, and through stable reconstitution of Tak1-deficient mouse embryo fibroblasts (MEFs) with the Lys209→Arg mutant of TAK1 they found that K63-linked polyubiquitin was only required for the rapid and transient activation of NF-κB in response to IL-1β. Surprisingly, polyubiquitination of TAK1 was completely dispensable for the late activation of NF-κB, which demonstrated for the first time that early, transient activation of NF-κB is mechanistically distinct from late activation of NF-κB.

What, then, is the mechanism of late activation of NF-κB? Several groups have previously demonstrated that both the RING domain and the zinc fingers of TRAF6 are important for the activation of NF-κB (911). Because the early activation was dependent on the RING domain and the subsequent K63-linked polyubiquitination of TAK1, the authors turned to the zinc fingers to look for clues about the late activation mechanism. They systematically mutated the zinc fingers of TRAF6 and found that mutation of the fifth finger (T6mZ5) selectively abrogated the late activation of NF-κB without interfering with its early activation. These results further confirmed that the early and late mechanisms of NF-κB activation were distinct, could be uncoupled from each other, and were controlled by different domains of TRAF6. Yamazaki et al. term these two pathways (based on their TRAF6 domain requirements) the “RING pathway” and the “zinc pathway,” respectively (Fig. 1). These two pathways provide a new conceptual framework to understand how the activation of NF-κB is temporally controlled by proinflammatory cytokines. Furthermore, this framework may also apply to other factors and pathogenic stimuli that use TRAF6 as a signaling component, such as those induced by the large family of Toll-like receptors (TLRs) that recognize pathogen-associated molecular patterns (12).

The investigation of the K63-linked polyubiquitin of TAK1 also led the authors to make a surprising discovery regarding the mechanism of IL-1β–induced activation of NF-κB. Previous genetic studies have demonstrated that mitogen-activated protein kinase (MAPK) kinase kinase 3 (MEKK3), a member of the family of kinases to which TAK1 belongs, is also required for the activation of NF-κB by IL-1β (13); however, its specific functional role was unclear (1417). Experiments with Mekk3-deficient MEFs showed that this kinase is required for both early and late activation of NF-κB (analyzed up to 60 min) (13). Yamazaki et al. found that wild-type TAK1 associated with MEKK3 but that the K209R mutant of TAK1, which cannot undergo K63-linked polyubiquitination, failed to do so, which suggests that K63-mediated polyubiquitination is required for formation of the TAK1-MEKK3 complex. Moreover, MEKK3 was required for activation of TAK1 in response to IL-1β, indicating that MEKK3 functions as an upstream kinase of TAK1 to mediate the early RING pathway. Because MEKK3 does not harbor a ubiquitin-binding domain, the authors suggest the involvement of a MEKK3-associated ubiquitin-binding subunit (possibly p62) that recruits this kinase to ubiquitin-modified TAK1 to activate it. Consistent with the above genetic evidence, the authors found that the later zinc pathway also depended on MEKK3, although its specific role in mediating downstream activation of IKK remains to be elucidated (Fig. 1).

The study by Yamazaki et al. provides a potential paradigm shift in our understanding of the role of K63-linked polyubiquitination and how this modification participates in the temporal regulation of the activation and function of NF-κB. It also raises a number of intriguing questions. First, how does MEKK3 act in two seemingly independent mechanisms of activation of IKK and NF-κB? Although these mechanisms are separable based on their specific domain requirements in TRAF6, are they nevertheless biochemically linked? For example, does the zinc pathway originate from the same TRAF6-MEKK3-TAK1 complex that mediates the RING pathway after removal of the K63-linked polyubiquitin chain from TAK1 by a deubiquitinase (DUB), such as CYLD or A20 (1821) (Fig. 1)? Second, how is the signaling that is mediated by the RING and zinc pathways regulated over time? Although removal of K63-linked polyubiquitin chains by DUBs could provide a basis for rapid termination of RING pathway signals, what factors sustain the zinc pathway for longer durations? Third, what is the relative biological importance of the RING and zinc pathways? Analysis of the selective induction of the expression of NF-κB–dependent genes in response to IL-1β stimulation indicated that both the RING and zinc pathways were important, with the former being more important for the production of IL-6. However, analysis of animal models selectively defective for each of the pathways may be required to reveal the physiological functions of these pathways. Finally, how can the role of MEKK3 as a kinase of TAK1 in the RING pathway be integrated into or rectified with an alternative model in which free, “unanchored” K63-linked polyubiquitin chains directly activate TAK1 by promoting its TAB2-dependent multimerization and trans-autophosphorylation (22)? Given the rapid pace at which new discoveries are emerging, the jury is still out, and you will be the judge.


Stay Connected to Science Signaling

Navigate This Article