Pococurante (Poc): An ENU-Induced Missense Mutation of MyD88 with Selective Effects on TLR Signal Transduction.

Toll-like receptors (TLRs) are key sensors of the innate immune system, and individual TLRs respond to specific molecules derived from microbes. MyD88 is a Toll/Interleukin-1/Resistance (TIR) domain-containing adaptor protein required for signaling by all TLRs except TLR3. While the structural basis of association between MyD88 and TIR-domain receptors is obscure, MyD88-deficient mice show no responses to bacterial flagellin, peptidoglycan (PGN), lipoteichoic acid (LTA), bacterial lipopeptides such as PAM₂CSK₄, PAM₃CSK₄ and R- or S-MALP-2, DNA bearing unmethylated CpG dinucleotides (CpG DNA), or Resiquimod (RSQ). Using germline ENU mutagenesis, we have produced a large number of phenotypic variants that have abnormal TLR signaling. We now report the identification of a new mutation called Pococurante (Poc), originally detected in screening because macrophages from this mouse showed no response to the tri-acylated lipopeptide PAM₃CSK₄, the di-acylated lipopeptide S-MALP-2, LTA, CpG DNA, RSQ, and a markedly reduced response to LPS: the ligands for TLRs 2/1, 2/6, 9, 7 and 4 respectively. They also had no response to interleukin-1, a cytokine that signals by way of a MyD88-dependent TIR domain receptor. However, Poc mice showed a normal response to PGN, as well as R-MALP-2 and PAM₂CSK₄ lipopeptides. The latter three ligands are sensed in a TLR2-dependent, MyD88-dependent fashion. The Poc phenotype was ascribed to a point mutation of MyD88 affecting a surface residue (I179N). Because the mutation is discriminatory, permitting MyD88 to carry a signal from some TIR domain receptors but not others, we infer that it resides at the receptor:adaptor signaling interface. A new model of TIR receptor:adaptor interaction is proposed on the basis of docking studies that take account of the Poc phenotype, made using the protein-protein docking program SURFDOCK. We note that S-MALP-2 is dependent upon TLR2/6 heterodimers, while PAM₃CSK₄ sensing depends upon TLR2/1 heterodimers. Since the Poc mutation forbids detection of both these ligands while it allows detection of PAM₂CSK₄ and R-MALP-2, it may be inferred that TLR2 signal transduction entails greater structural diversity than was previously supposed. The involvement of TLR2 homodimers, or the incorporation of subunits yet unknown into the receptor complex, cannot be excluded.