Structure and Function of the Pseudokinases IRAK2 and IRAK3 and their Potential as Drug Targets

  • Sven Manfred Lange

Student thesis: Doctoral ThesisDoctor of Philosophy


The Toll-like receptors (TLRs) and interleukin-1 receptors (IL-1R) are critical components of the innate immune system that recruit the adapter protein MyD88 (myeloid differentiation primary response gene 88), which serves as an assembly platform for members of the interleukin-1 receptor associated kinase (IRAK) family to form an oligomeric signalling complex called the Myddosome. Subsequently, IRAK kinases and pseudokinases recruit and activate a number of E3 ubiquitin ligases that trigger downstream signalling events to induce the production of pro- and anti-inflammatory mediators. The emergence of crystal structures and development of specific inhibitors of the catalytically active IRAK family members, IRAK1 and IRAK4, have advanced our understanding of their roles in the TLR/IL-1R signalling pathway. However, there is no structural or mechanistic information about the two IRAK pseudokinases, IRAK2 and IRAK3. For this reason, I focused my research on the structural characterisation of the IRAK pseudokinases, to decode their roles within MyD88-dependent signalling. Chapter 3 is focused on the pseudokinase domains of IRAK2 and IRAK3. Chapter 4 describes the identification and characterisation of small molecules that bind to IRAK2 and IRAK3, to evaluate their potential as drug targets. Chapter 5 centres on the interaction of the E3 ubiquitin ligase TRAF6 with IRAK2 and IRAK3.

I discovered that the kinase domain of IRAK2 is monomeric in solution, while IRAK3 is in a concentration-dependent equilibrium of monomeric and dimeric states. ATP-binding assays revealed that IRAK2 binds ATP in a Mg-dependent fashion, while IRAK3 does not bind ATP, but binds to the ATP-competitive pankinase inhibitor staurosporine.

I present the first crystal structure of the pseudokinase domain of human IRAK3, which reveals a pseudoactive conformation and provides support for IRAK3’s inability to bind ATP. IRAK3 forms a hexamer in the crystal via two alternating pseudosymmetric homo-dimer interfaces: The αC-to-αC interface, which centres on the αC-helices of IRAK3, and the αG-to-αG interface with central αG-helices. Structure-guided point-mutations in the αC-to-αC interface identified this interface to form the in-solution dimer of IRAK3. Interestingly, I found that a higher-order assembly of IRAK3 pseudokinase domains with alternating αC-to-αC and αG-to-αG interfaces would create a helical IRAK3 filament.

Strikingly, I identified a third putative interface on the IRAK3 surface in a conservation analysis of vertebrate IRAK3 sequences. The third interface is centred around the αEF-helix of IRAK3 and closely resembles the αEF-to-αEF interface of the previously reported IRAK4 homo-dimer. Mutations in IRAK3 that are linked to the pathogenesis of early-onset persistent asthma lie in the putative αEF-interface of IRAK3. Intriguingly, the three distinct dimerisation interfaces on the surface of the pseudokinase domain of IRAK3 do not overlap and allowed for the creation of a higher-order assembly model of IRAK3 and IRAK4 molecules. The model resembles a potential IRAK3-mediated inhibited state of the IRAK4 kinase domain, which would be a novel mechanism of kinase inhibition by a pseudokinase oligomer. Furthermore, these findings demonstrate for the first time that the ability of the Myddosome to signal may be tightly regulated by oligomerisation of the kinase and pseudokinase domains of the IRAKs.

In collaboration with Janssen Pharmaceutica, I performed a high-throughput assay to screen for small molecule binders of the IRAK pseudokinases. I identified and started to characterise twelve small molecules that bind to the IRAK3 pseudokinase domain with nanomolar affinities.

Lastly, I detected a highly conserved helical region in the C-terminal domains of IRAK1 and IRAK2, which might be the first clue for a function of the C-terminal domains of the IRAKs in addition to their role in TRAF6-binding. I investigated the interaction between TRAF6 and the C-terminal domains of IRAKs and developed a high-throughput HTRF assay for the identification of protein-protein interaction inhibitors to disrupt this interaction. Together, this work lays the foundation to explore the IRAK pseudokinases as potential drug targets for the treatment of innate immune diseases in the future.
Date of Award2020
Original languageEnglish
SupervisorPhilip Cohen (Supervisor), Yogesh Kulathu (Supervisor) & Marina Nelen (Supervisor)


  • pseudokinase
  • IRAK
  • TRAF6
  • Drug discovery
  • crystallography
  • Structural biology
  • Immunity

Cite this