The Architecture of the TIR Domain Signalosome in the Toll-like Receptor-4 Signaling Pathway

Abstract

Activated Toll-like receptors (TLRs) cluster in lipid rafts and induce pro- and anti-tumor responses. The organization of the assembly is critical to the understanding of how these key receptors control major signaling pathways in the cell. Although several models for individual interactions were proposed, the entire TIR-domain signalosome architecture has not been worked out, possibly due to its complexity. We employ a powerful algorithm, crystal structures and experimental data to model the TLR4 and its cluster. The architecture that we obtain with 8 MyD88 molecules provides the structural basis for the MyD88-templated myddosome helical assembly and receptor clustering; it also provides clues to pro- and anti-inflammatory signaling pathways branching at the signalosome level to Mal/MyD88 and TRAM/TRIF pro- and anti-inflammatory pathways. The assembly of MyD88 death domain (DD) with TRAF3 (anti-viral/anti-inflammatory) and TRAF6 (pro-inflammatory) suggest that TRAF3/TRAF6 binding sites on MyD88 DD partially overlap, as do IRAK4 and FADD. Significantly, the organization illuminates mechanisms of oncogenic mutations, demonstrates that almost all TLR4 parallel pathways are competitive and clarifies decisions at pathway branching points. The architectures are compatible with the currently-available experimental data and provide compelling insights into signaling in cancer and inflammation pathways.

Figure 1. Toll-like receptor pathway (adapted from literature121519), in traditional node-and-edge representation, where nodes represent proteins and edges represent interactions between proteins.

TLR pathway is complicated and has many branches. Stimulation of TLRs propagate the signal through two parallel paths: MyD88-dependent path (green), which leads to production of pro-inflammatory cytokines, and TRIF-dependent path (orange), which gives rise to transcription of antiviral proteins—interferons—and anti-inflammatory cytokine IL-10. MyD88-mediated pathway also has three branches, namely TRAF6- (green), TRAF3- (orange) (downstream of endosomal TLRs), and FADD-dependent (pink) downstream pathways. For space limitation, we showed TLRs on endosomal membrane as monomers, but they also dimerize upon stimulation.

Figure 2. 3D schematic view of TIR-domain signalosome, myddosome and TLR clustering.

It is known that TLRs cluster on lipid rafts, but they cannot tetramerize due to the steric hindrance of their ectodomains. Oligomerization of the downstream proteins may hold TLRs together. Here, all TIR domains are in dimer form, TLR4, Mal, and MyD88. A TLR4-dimer recruits two Mal-dimers, which in turn recruit four MyD88-dimers. In the myddosome complexes, there are six MyD88 molecules, four IRAK4 and four IRAK2. The box at the upper right corner shows the cartoon version of the model. The PDB_ID of the myddosome complex is 3mop: MyD88 death domains 3mopBCDE, IRAK4 death domains 3mopGHIJ, IRAK2 death domains 3mopKLMN. Pink circles are PI(4,5)P2, which are enriched in lipid rafts and N-terminal region of Mal associates with it.

Figure 3. TLR4 homodimer models.

(a) Back-to-back TLR4 dimer (BB). (b) Face-to-face TLR4 dimer (FF), which is very similar a previously suggested TLR4-dimer model and crystal structure of the dimeric TIR domain of IL-1RAPL (1t3g.pdb). (c) Another face-to-face dimer (FF2) in which BB-loops are in very close proximity. The box in the lower left corner shows the structural alignment of this TLR4-homodimer model with the one that is obtained by superimposition of TLR4 with TLR2 homodimer crystal structure (1o77_CD), (146 of 276 residues with 0.73 RMSD by multiport). Cyan color is TLR4 TIR domain, red-labeled regions are BB-loops, and yellow spheres are C747 residues on each TLR4 TIR domain, which are suggested to be involved in the interface. The dark blue dimer in the box is TLR4-dimer, which is obtained by superimposition with TLR2 dimer (1o77_CD).

Figure 4

Interaction models of Mal-monomer (a,c) and Mal-dimer (b,d) with BB and FF TLR4-homodimer models. Yellow protein is Mal and green spheres show the proposed interface residues of Mal (R184, A185, Y187), none of which are at the correct site in the monomeric-Mal-TLR interaction model. However, if dimerization of Mal is also taken into account, it is seen that both monomers are in contact with TLR4, one of which has the interface residue at the correct site.

Figure 5. Possible TIR domain signalosome models for FF TLR4-dimer.

(a) Interaction model of monomeric-MyD88 with TLR4 and Mal dimers. (b,c) MyD88-dependent TIR-domain signalosome models for FF TLR4-dimer. All proteins are in dimer form, including TLR4, Mal, and MyD88. It is known that dimeric MyD88 has higher affinity to stimulated TLRs due to their extended interfaces. In line with this, models (b,c) show that the second MyD88 of the MyD88-dimer is very close to TLR4. Especially in part-c, one of the MyD88 molecules in the dimer is bound to Mal, and the other is bound to TLR4. We obtained these complexes by superimposition of the binary interaction models of TLR4-TLR4, TLR4-Mal, Mal-Mal, Mal-MyD88 and MyD88-MyD88.

Figure 6. MyD88- and TRIF- dependent downstream TLR pathways are mutually exclusive.

(a) TRAM-homodimer has a steric clash with Mal-homodimer when superimposed to BB TLR4-homodimer model, and thus they are mutually exclusive: either Mal or TRAM homodimers can bind to TLR4 at any time. TRAM and Mal interactions are mutually exclusive in BB TLR4-homodimers and this is in line with the findings of several studies. This indicates that MyD88-dependent pro-inflammatory and TRIF-dependent anti-inflammatory pathways are competitive. (b) TRAM-homodimer does not overlap with Mal-homodimer when superimposed to FF TLR4-homodimer model. (c) MyD88 overlaps with TRIF on TLR4: the FF TLR4-homodimer model has steric clashes of MyD88 and TRIF when superimposed Mal-MyD88 and TRAM-TRIF. Red box indicates the location of steric clash.

Figure 7. MyD88 interaction models with the downstream orchestrators reveal that the three parallel downstream paths are competitive.

(a) TRAF6 (1lb5:A) and TRAF3 (1fll:A) binds to almost completely overlapping interfaces on MyD88 DD (3mop:F), thus they are mutually exclusive. (b) IRAK4 (3mop:J) and FADD (2gf5:A) bind to overlapping interfaces on MyD88 DD (3mop:F), thus they compete to bind to MyD88. MyD88-IRAK4 interaction is not PRISM prediction, where the crystal structure of the complex is available (3mop:FJ). Red box indicates the location of steric clash.

Figure 8. Parallel downstream pathways of TLRs, which lead to distinct outcomes, are mutually exclusive.

Green arrows shows that TRIF- and MyD88-dependent paths cannot be activated simultaneously due to shared binding sites on TLR4-dimer or steric hindrance. Blue arrows demonstrate that TRAF6 and TRAF3 bind to overlapping interfaces on MyD88 DD (downstream of endosomal TLRs). Pink arrows shows that IRAK4 and FADD will have steric clash when they bind to MyD88 at the same time. The three branches of TLR pathway, namely pro-inflammatory, interferon and anti-inflammatory, and apoptotic paths are mutually exclusive.