664 Biochemical Society Transactions (2003) Volume 31, part 3 Toll-like receptor 4 signalling: new perspectives on a complex signal-transduction problem S.N. Vogel 1 and M. Fenton Department of Microbiology and Immunology, and Department of Medicine, University of Maryland, Baltimore, 655 W. Baltimore Street 13-009, Baltimore, MD 21201, U.S.A. Abstract We previously reported that Toll-like receptor-2 (TLR2) agonists induce expression of a more limited repertoire of pro-inflammatory genes than TLR4 agonists. Murine macrophages stimulated with the TLR4 agonist, Escherichia coli lipopolysaccharide, induced signal transducer and activator of transcription 1 (‘STAT1’) tyrosine phosphorylation that was secondary to the autocrine/paracrine action of interferon (IFN)-β , an immediate early gene. In contrast, TLR2 agonists failed to activate IFN-β gene expression. TLR4-induced IFN-β mRNA was found to be MyD88- and PKR (double-stranded RNA-dependent protein kinase)-independent, but TIRAP (Toll/interleukin-1 receptor domain-containing adapter protein)/Mal (MyD88-adapter-like)-dependent. In the present paper, we outline the recent controversy over the role of TIRAP/Mal in TLR2 and TLR4 signalling in the context of the current molecular tools used for such studies. Collectively, our findings provide the first mechanistic basis for differential patterns of gene expression activated by TLR4 and TLR2 agonists. Introduction Lipopolysaccharide (LPS) is among the most inflammatory of bacterial products. It is a common integral structural component of the outer membrane of Gram-negative bacteria. A primary physiological consequence of LPS– macrophage interaction is the release of an amazing spectrum of soluble mediators, many of which are synthesized de novo. Specifically, a variety of cytokines, chemokines, lipid-derived mediators and oxygen-derived antibacterial products are among those released by LPS-stimulated macrophages. The intensity and longevity of inflammatory mediator production dictates whether a response will be protective or whether an overwhelming inflammatory response will ensue, as is often seen in Gram-negative sepsis. All of our studies, in vivo or in vitro, have focused on understanding how to regulate this system to augment or dampen the response. The follow- ing will summarize our findings and postulate mechanisms of Toll-like receptor (TLR) signalling, leading to gene expression. Key words: interferon, interleukin, MyD88, MyD88-adapter-like (Mal), Toll-like receptor, TIR domain-containing adapter protein (TIRAP). Abbreviations used: TLR, Toll-like receptor; LPS, lipopolysaccharide; STAT1, signal transducer and activator of transcription 1; PKR, double-stranded RNA-dependent protein kinase; IL-1, interleukin-1; TIR, Toll/IL-1 receptor; TIRAP, TIR domain-containing adapter protein; Mal, MyD88-adapter-like; RT, reverse transcription; TNF-α, tumour necrosis factor-α; Cox-2, cyclo-oxygenase-2; IFN, interferon; IP-10, IFN-inducible protein-10; ICSBP, IFN con- sensus sequence-binding protein; NF-κB, nuclear factor-κB; iNOS, inducible nitric oxide synthase; Pam3Cys, S-[2,3-bis(palmitoyloxy)-(2-R,S)-propyl]-N-palmitoyl-(R)-Cys-(S)-Ser-Lys4 - hydroxytrihydrochloride; MCP-5, monocyte chemoattractant protein 5; MAP, mitogen-activated protein; DN, dominant negative; CpG, 2 ′ -deoxyribo(cytidine-phosphate-guanosine); IRF-3, interferon regulatory factor 3. 1 To whom correspondence should be addressed (e-mail svogel@som.umaryland.edu). LPS-induced gene expression in macrophages requires interaction of multiple proteins within an ‘LPS receptor complex’ Using macrophages from mice with natural or targeted mutations, we found that co-ordinate receptor utilization is required for optimal LPS-induced gene expression [1,2]. For example, we have used peritoneal exudate macrophages from C3H/HeJ mice, which have a point mutation in the intracytoplasmic domain of the TLR4 receptor that precludes normal LPS signalling [3,4], as well as mice with a targeted mutation in TLR4 [5]. Using reverse transcription (RT)- PCR with Southern blot analysis or Northern blot analysis, we have shown that TLR4 is necessary for expression of all LPS-inducible genes examined, e.g. tumour necrosis factor-α (TNF-α), interleukin (IL)-12 p40, IL-12 p35, cyclo- oxygenase-2 (Cox-2), interferon (IFN)-inducible protein-10 (IP-10) and IFN consensus sequence-binding protein (ICSBP). Increasing the concentration of LPS to μg/ml levels fails to overcome this defect. Macrophages derived from mice with a targeted mutation in CD14, a soluble or glycosylphosphatidylinositol-anchored glycoprotein, are also very deficient in LPS-induced gene expression at low concentrations of LPS. However, in contrast with macro- phages derived from TLR4-deficient mice (e.g. C3H/HeJ or TLR4 knockout mice), most genes can be induced in CD14 knockout macrophages by increasing the concentration of LPS [6]. However, certain genes, such as those for IP-10 and ICSBP, are extremely dependent upon the presence of CD14 for induction, since increasing the concentration of LPS does C  2003 Biochemical Society