An investigation of the activation of protein kinase complexes in the MyD88 signalling network

  • Jiazhen Zhang

Student thesis: Doctoral ThesisDoctor of Philosophy


The TAK1 and canonical IKK complexes are the two master protein kinases of the innate immune system that control the production of inflammatory mediators, but the mechanisms by which they are activated in this system are still unclear. In this thesis, I present the research I have carried out to solve these problems.
The IKKb component of the canonical IKK complex is required to activate the transcription factors NF-kB and IRF5 and the protein kinase Tpl2, but how IKKβ itself is activated in vivo is still unclear. It was found to require phosphorylation by one or more ‘upstream’ protein kinases in some reports, but by autophosphorylation in others. In the first part of this thesis, I describe my work that has resolved this controversy by demonstrating that the activation of IKK induced by IL-1 (interleukin-1) or TNF (tumour necrosis factor) in embryonic fibroblasts, or by ligands that activate Toll-like receptors in macrophages, requires two distinct phosphorylation events: first, the TAK1 catalysed phosphorylation of Ser177 and, secondly, the IKKb-catalysed autophosphorylation of Ser181. The phosphorylation of Ser177 by TAK1 is a priming event required for the subsequent autophosphorylation of Ser181, which enables IKKk to phosphorylate exogenous substrates. I also present genetic evidence which indicates that the IL-1-stimulated, LUBAC (linear ubiquitin chain assembly complex)-catalysed formation of Met1-linked/linear ubiquitin (Met1-Ub) chains and their interaction with the NEMO (NF-kB essential modulator) component of the canonical IKK complex permits the TAK1-catalysed priming phosphorylation of IKKb at Ser177 and IKKa at Ser176. These findings may be of general significance for the activation of other protein kinases.
The activation of the TAK1 complex by inflammatory stimuli is thought to be triggered by the binding of Lys63-linked ubiquitin chains to the TAB2 or TAB3 components of the TAB1-TAK1-TAB2 and TAB1-TAK1-TAB3 complexes. In the second part of the thesis I tested whether this broadly accepted model was correct by knocking out the genes encoding TAK1 and its regulatory subunits TAB1, TAB2 and TAB3 by CRISPR/Cas9 gene-editing technology, alone and in combination, in an IL-1 receptor expressing human cell line. These genetic studies led me to discover that the IL-1-dependent activation of TAK1 occurs by two different mechanisms. The first, involves the previously described interaction of Lys63-linked ubiquitin chains with TAB2 and TAB3, while the second can take place in the complete absence of TAB2 and TAB3. The second mechanism, which involves activation of the TAB1-TAK1 heterodimer is more transient than the first, but is sufficient for the IL-1-dependent transcription of immediate early genes (A20, IkBa). I show that the activation of the TAB1-TAK1 complex requires the expression of the E3 ubiquitin ligase TRAF6 and the TRAF6-generated formation of Lys63-linked ubiquitin chains, which leads to the phosphorylation of TAK1 at Thr187 and activation. However, neither TAB1 nor TAK1 bind directly to Lys63-linked ubiquitin chains. I identify one novel IL-1-dependent phosphorylation site on TAB1 and two on TAK1 and propose that Lys63-linked ubiquitin chains activate an as yet unidentified protein kinase, which phosphorylates one or more of the novel phosphorylation sites on the TAB1-TAK1 heterodimer inducing a conformational change that permits TAK1 to autophosphorylate Thr187.
Date of Award2017
Original languageEnglish
SupervisorPhilip Cohen (Supervisor)


  • TAK1
  • Canonical IKK complex
  • MyD88
  • Kinase

Cite this