A conserved surface on Toll-like receptor 5 recognizes bacterial flagellin

Abstract

The molecular basis for Toll-like receptor (TLR) recognition of microbial ligands is unknown. We demonstrate that mouse and human TLR5 discriminate between different flagellins, and we use this difference to map the flagellin recognition site on TLR5 to 228 amino acids of the extracellular domain. Through molecular modeling of the TLR5 ectodomain, we identify two conserved surface-exposed regions. Mutagenesis studies demonstrate that naturally occurring amino acid variation in TLR5 residue 268 is responsible for human and mouse discrimination between flagellin molecules. Mutations within one conserved surface identify residues D295 and D367 as important for flagellin recognition. These studies localize flagellin recognition to a conserved surface on the modeled TLR5 structure, providing detailed analysis of the interaction of a TLR with its ligand. These findings suggest that ligand binding at the beta sheets results in TLR activation and provide a new framework for understanding TLR-agonist interactions.

Figure 1.

Mouse TLR5 detects most flagellins better than human TLR5. (A) Immunoblot of CHO cells stably expressing vector alone or human (h) or mouse (m) TLR5. 100 μg of cellular cytoplasmic extracts were loaded per lane, and TLR5 expression was detected by immunoblotting for the HA epitope tag. Equal loading was verified by immunoblotting for β-tubulin. Kilodalton values are shown. (B–F) Shown is the percent fold induction of NF-κB luciferase activity for flagellin purified from each indicated bacterial species relative to maximal stimulation achieved with S. typhimurium flagellin for cells expressing either human (continuous line) or mouse (dashed line) TLR5 at the indicated flagellin doses. Data are representative of at least three independent experiments. Error bars represent the mean ± SD. (G) Table listing the effective flagellin concentrations needed to achieve half maximal activation of TLR5 (EC₅₀). p-values were calculated using a two-tailed Student's t test.

Figure 2.

The extracellular domain of TLR5 is responsible for flagellin recognition. (A and B) Dose-response curves of CHO cells transfected with human (A) and mouse (B) TLR5 to wild-type flagellin from S. typhimurium, point mutants I411A and R90A, and the double mutant R90A/I411A. Fold induction is relative to the maximal induction seen for wild-type flagellin. (C) Table showing a linear schematic of TLR5, with mouse TLR5 amino acid numbers of the domain boundaries shown above the molecule. EC, extracellular domain; TM, transmembrane domain; TIR, Toll/IL-1 receptor domain. (D) Immunoblot of CHO cells stably expressing human (h) and mouse (m) TLR5 or the extracellular domain TLR5 chimeras (hm and mh). 100 μg of cellular cytoplasmic extracts were loaded per lane, and TLR5 expression was detected by immunoblotting for the HA epitope tag. Equal loading was verified by immunoblotting for β-tubulin. Kilodalton values are shown. (E and F) Dose-response curves of the hm (E) and mh (F) chimeric TLR5 receptors to wild-type flagellin from S. typhimurium, point mutants R90A and I411A, and the double mutant R90A/I411A.

Figure 3.

The central 228 amino acids of the TLR5 extracellular domain confer species-specific flagellin recognition. (A) Table showing a linear schematic of TLR5, with amino acid numbers of the domain boundaries shown above the molecule. EC, extracellular domain; TM, transmembrane domain; TIR, Toll/IL-1 receptor domain. (B) Immunoblot of CHO cells stably expressing human (h) and mouse (m) TLR5 or the extracellular domain TLR5 chimeras (hmh, mhh, mhmm, and mmhm). 100 μg of cellular cytoplasmic extracts were loaded per lane, and TLR5 expression was detected by immunoblotting for the HA epitope tag. Equivalent loading was verified by immunoblotting for β-tubulin. Kilodalton values are shown. (C–F) Dose-response curves of the hmh (C), mhh (D), mmhm (E), and mhmm (F) chimeric TLR5 receptors to wild-type flagellin from S. typhimurium, point mutants R90A and I411A, and the double mutant R90A/I411A. The percent fold induction is relative to the maximal induction seen for wild-type flagellin. (G) Immunoprecipitation of chimeric TLR5 molecules with the wild type or the R90A flagellin mutant.

Figure 4.

Model of the TLR5 extracellular domain. (A) Ribbon representations (far left and far right) and molecular surface representations (middle left and middle right) of the best model of the TLR5 ectodomain (left) and the structure of flagellin (PDB lucu; right). The ribbon is colored sequentially from the N terminus (blue) to the C terminus (red). Amino acids important for flagellin recognition are shown on the TLR5 model in red. Amino acids on flagellin previously determined to be important for TLR5 recognition are shown in red on the flagellin structure (reference 5). (B) Ribbon representations (left) and molecular surface representations (middle and right) of the best model of the TLR5 ectodomain, oriented ∼90° to the view in A. Views 180° apart are shown in the top and bottom rows. The ribbon is colored sequentially from the N terminus (blue) to the C terminus. Molecular surface representations are colored by conservation (center: residues ≥90% similar among TLR5 sequences are green) or by electrostatic potential (right: positive, blue; negative, red). The conserved concavity and lateral patch regions are indicated with a dotted white oval and bracket, respectively. Positions of residues important for flagellin recognition are indicated with arrows. (C) Space-filling representation of the TLR5 model. Residues of the conserved concavity (white) and residues that differ between human and mouse TLR5 in the central 228–amino acid region, as well as 407 and 408 (red), are highlighted. Amino acids surrounding the concavity that were mutated are indicated with arrows.

Figure 5.

P268 confers species-specific recognition of flagellin point mutants. (A) Immunoblot of CHO cells stably expressing human (h) and mouse (m) TLR5 or the extracellular domain TLR5 mutants (mQ407P/M408D, mL292R/Q293H, mP268A, or hA268P). 100 μg of cellular cytoplasmic extracts were loaded per lane, and TLR5 expression was detected by immunoblotting for the HA epitope tag. Equal loading was verified by immunoblotting for β-tubulin. Kilodalton values are shown. (B–E) Dose-response curves of the mQ407P/M408D (B), mL292R/Q293H (C), mP268A (D), and hA268P (E) mutant TLR5 receptors to wild-type flagellin from S. typhimurium, point mutants R90A and I411A, and the double mutant R90A/I411A.

Figure 6.

The conserved concavity on TLR5 interacts with flagellin. (A) Immunoblot of CHO cells stably expressing mouse TLR5 (m) or the TLR5 concavity point mutants (mD295A, mS297A, and mD367A). 100 μg of cellular cytoplasmic extracts were loaded per lane, and TLR5 expression was detected by immunoblotting for the HA epitope tag. Equal loading was verified by immunoblotting for β-tubulin. Kilodalton values are shown. (B–E) Dose-response curves for wild-type flagellin (B) from S. typhimurium and flagellin point mutants R90A (C), I411A (D), and R90A/I411A (E) for each TLR5 point mutant.