How Toll-like receptors and Nod-like receptors contribute to innate immunity in mammals

Transcription

1 Proceedings How Toll-like receptors and Nod-like receptors contribute to innate immunity in mammals Jörg H. Fritz 1, Stephen E. Girardin 2 1 Immunité Innée et Signalisation Laboratoire, and 2 Pathogénie Microbienne Moléculaire, Institut Pasteur, Paris, France Innate immune detection of pathogens relies on specific classes of microbial sensors called patternrecognition molecules (PRM). In mammals, such PRM include Toll-like receptors (TLRs) and the intracellular proteins NOD1 and NOD2, which belong to the family of Nod-like receptors (NLRs). Over the last decade as these molecules were discovered, a function in innate immunity has been assigned for the majority of them and, for most, the microbial motifs that these molecules detect were identified. One of the next challenges in innate immunity is to establish a better understanding of the complex interplay between signaling pathways induced simultaneously by distinct PRMs and how this affects tailoring first-line responses and the induction of adaptive immunity to a given pathogen. Keywords: Toll-like receptors, Nod-like receptors, innate immunity INTRODUCTION Innate immunity is the most ancient system of defense against pathogens found in animals and plants. In vertebrates, an additional defense system known as adaptive immunity exists, which relies on somatic re-arrangement of genes to give rise to tailor-made highly specific antigen receptors. 1 In contrast, innate immunity relies on a limited set of pattern recognition molecules (PRMs) that detect invariant structures of microbes termed pathogenassociated molecular patterns (PAMPs), such as lipopolysaccharide (LPS), peptidoglycan (PG), lipoproteins, CpG DNA or viral RNA. 2 For most of these microbial motifs, it has been known for a long time that they can act as immune modulators when present in body fluids of arthropods or vertebrates. However, until recently, the molecules responsible for their detection in the host have remained unidentified. Less than a decade ago, Toll was the first PRM identified in Drosophila; soon after, a Received 17 December 2004 Revised 28 March 2005 Accepted 9 May 2005 Correspondence to: Stephen E. Girardin, Pathogénie Microbienne Moléculaire, Institut Pasteur, rue du Dr Roux, Paris, France. Tel: ; Fax: ; girardin@pasteur.fr Journal of Endotoxin Research, Vol. 11, No. 6, 2005 DOI / X76850 mammalian family of homologous proteins was characterized and termed Toll-like receptors (TLRs). 3 The characterization of microbial motifs detected by each member of the TLR family (so far, 10 in humans and 11 in mice) is the subject of intense investigation. 4 In addition to TLRs, an additional family of proteins called Nod-like receptors (NLRs) has been discovered, and the microbial motifs detected by two of their members, NOD1 and NOD2, have been characterized recently. 5 PRM FAMILIES Defining the minimal structures of microbial molecules required for fully-fledged innate immune signaling is of extreme importance for gathering knowledge of downstream pathways induced by individual PAMP PRM engagement. The specificities for many PRMs have been Abbreviations: PRM, pattern-recognition molecules; TLR, Toll-like receptor; NLR, Nod-like receptor; PAMP, pathogen-associated molecular patterns; NACHT, NAIP, CIITA, HET-E, TP-1; CARD, caspase recruitment domain; LRR, leucine-rich repeat; PYD, pyrin domain; BIR, baculovirus inhibitor of apoptosis protein repeat; NOD, nucleotide-binding oligomerization domain; NAIP, neuronal apoptosis inhibitor protein; CIITA, major histocompatibility complex (MHC) class II transactivator; HET-E, plant het gene product involved in vegetative incompatibility; TP1, telomerase-associated protein 1. W. S. Maney & Son Ltd

2 How Toll-like receptors and Nod-like receptors contribute to innate immunity in mammals 391 deciphered; for some, a good deal is known about how they signal. 6 A direct consequence of this is the use of various chemically-defined PAMPs in models of cancer and allergy to enhance acquired immune responses for prophylactic and therapeutic vaccination. 7 However, upon interaction of microbes with host cells, a multitude of diverse PAMP PRM recognition events takes place. Although important knowledge has accumulated over the past few years about each individual PRM (and the pathways directly downstream), little is know about how these cascades influence each other. Functionally, PRMs can be divided into four types: (i) humoral components circulating in the plasma such as acute phase response proteins or complement that opsonize microbes or derivatives thereof marking them for recognition by other receptors; (ii) endocytic receptors expressed on the cell surface like scavenger receptors (SRs) or lectin receptors which directly can bind structures on microbial surfaces; (iii) opsonic phagocytic receptors (Fc and complement receptors [CR]) that recognize antibody and complement-coated particles; and (iv) PRMs such as TLRs and NLRs, which posses the intrinsic capability of converting the information gleaned from recognition of a pathogen to elicit specific downstream signaling events. TLRs Most PRMs do not possess cytoplasmic motifs to activate signaling events and only with the description of TLRs were insights provided about how innate immune activation occurred in response to PAMPs. Several PRMs such as TLR-1, TLR-2, TLR-4, TLR-5, TLR-6, TLR-9 and TLR-11 have been implicated in the sensing of bacterial molecules. 6 While TLR-5 has been shown to mediate the detection of flagellin, 8 TLR-9 has been reported to be an essential component in the sensing of unmethylated CpG motifs of bacterial DNA. 9 In a recent report, the specific sensing of uropathogenic bacteria has been associated with TLR-11, 10 although the exact microbial molecules essential for this responses are yet unknown. TLR-4 is an essential signaling molecule upon LPS detection, whereas TLR-2 has been reported to be engaged in transducing downstream effects of several molecular components of bacteria (PG, macrophage-activating lipopeptide 2 [MALP-2], lipoteichoic acid (LTA), di- and tri-acylated bacterial lipopeptides [Pam 2 and Pam 3, respectively]) and fungi (β-glucans) as reviewed by Underhill. 11 In addition, heterodimerization of TLR has been reported to be responsible for differential recognition of PAMPs such as the distinction of di- and tri-acylated lipopeptides by TLR-1/2 and TLR-2/6 heterodimers, respectively Until recently, studies linking PG with Table 1. PRMs and their activating PAMPs PRM Identified PAMP TLR-1/TLR-2 Tri-acyl lipopeptides TLR-2 Zymosan, LTA TLR-3 dsrna TLR-4 LPS, Taxol TLR-5 Flagellin TLR-6/TLR-2 Di-acyl lipopeptides TLR-7 Imidazoquinoline, Loxoribine TLR-8 ssrna TLR-9 Unmethylated CpG DNA TLR-10? TLR-11 Uropathogenic bacteria Nod1 GM-Tri DAP Nod2 MDP (GM-Di), M-TriLys TLR-2 recognition have relied mainly on the use of commercial Staphylococcus aureus PG. Using highly purified PGs from eight different bacteria, we have shown that PGs are not sensed through TLR-2, TLR-1/2 or TLR-2/6 after removal of lipoproteins or lipoteichoic acids from Gram-negative and Gram-positive cell walls. 15 It is clear that certain TLRs induce specific signaling cascades (reviewed by Kaisho and Akira 16 ), but whether TLRs directly recognize PAMPs as some studies suggest or whether accessory molecules perform this function remains to be investigated in more detail. Furthermore, TLRs show a high degree of promiscuity making discrimination between micro-organisms less precise. 16 The diversity in TLR-activating components has prompted the suggestion that different TLRs may activate different downstream responses and that these tailor immune responses to specific organisms. However, it is important to note that the effect of TLR deletion on microbial infection phenotypes is often not severe. For example, Listeria monocytogenes can induce inflammatory signaling through TLR-2, but mice deficient in TLR-2 are not more susceptible to infection. 17 Similarly, Escherichia coli LPS is recognized by TLR-4, but mice deficient in TLR-4 do not show enhanced susceptibility to peritoneal or intravenous infection with this bacterium. 18 Redundancy in recognition might reflect these observations, thus being of importance to prevent microbes from easily subverting recognition. Multiple PAMP PRM recognition events further increase the possibilities for fine-tuning controlled first-line responses without damaging the host. Thus, to understand the regulation of magnitude and consequences of first-line host defences, the cellular localization of immune recognition and the positive as well as negative co-operation of PRMs have to be taken into account. Depending on the cell type and molecule, the localization of TLRs seems to be restricted to the cell surface or specialized cell organelles, especially the endolysosomal

3 392 Fritz, Girardin compartment. For example, TLR-2 is located on the plasma membrane of macrophages and stays bound to its ligands such as yeast zymosan even after internalization in the phagosome. 12 In order to recognise hypomethylated CpG motifs, endocytosis must occur so that the cell can signal through TLR-9 which has to be translocated from the ER to the lysosome. 19 TLR-4 is found on the surface of macrophages but in the Golgiapparatus of intestinal epithelial cells. 20 To avoid recognition and escape from defense, bacteria have developed strategies for invading and crossing the epithelium. 21 Upon arrival at the sub epithelial space, bacteria encounter locally resident, as well as newly infiltrated, professional phagocytic cells. However, bacteria use a variety of strategies to avoid engulfment and degradation by the endolysosomal compartment by phagocytes. 22 One might assume that the cytosolic location provides optimal protection from immune recognition and response. However, even the cytosol is equipped to detect the presence of microbes. Recently, a family of proteins (NOD1 and NOD2) was discovered that, based on initial observations, 23 led to the speculation that they are involved in intracellular bacterial detection and the initiation of the immune response, representing a means of cytosolic surveillance. NLRs: the example of Nod1 and Nod2 NLRs are characterized by three structural domains. 24 Located at the C-terminus, leucine-rich repeat (LRR) domains are able to sense PAMPs or other molecules. The second, intermediary nucleotide binding domain termed NACHT (NAIP, CIITA, HET-E, TP-1), which is an acronym for the founding family members of the NLR family, NAIP (neuronal apoptosis inhibitor protein), CIITA (major histocompatibility complex [MHC] class II transactivator), HET-E (plant het gene product involved in vegetative incompatibility), and TP-1 (telomerase-associated protein 1) is essential for the oligomerization of these molecules and is a prerequisite for the further transduction of the signal mediated by the third, N-terminal effector domain. This can be a pyrin domain (PYD), a caspase recruitment domain (CARD) or a baculovirus inhibitor of apoptosis protein repeat (BIR) domain. The first NLRs shown to be involved in innate immune sensing by recognizing specific microbial molecules were NOD1 and NOD2. 25 Both of these cytoplasmic PRMs detect PG, which is a component of the cell wall of most bacteria, with the exception of Mycoplasma and possibly Chlamydia spp. However, the molecular motifs required for stimulation of these molecules are different, stressing the fact that NOD1 and NOD2 have unique and non-overlapping specificities towards PG. 26 PG is the major constituent of the cell wall of Gram-positive bacteria, while in Gram-negative bacteria, it is found as a thin layer in the periplasmic space. Furthermore, PG is responsible for providing shape and mechanical rigidity to bacteria and is composed of glycan chains that contain alternating N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) sugars cross-linked to each other by short peptides (stem pepides). An important difference between Gram-positive and Gram-negative peptidoglycan resides in the nature of the third amino acid of the stem peptides. 27 In Gram-positive bacteria, this amino acid is commonly a lysine, whereas diaminopimelic acid (DAP) is found in most Gram-negative bacteria. The minimal structure detected by NOD1 is the dipeptide D-Glu-meso-DAP, in which meso-dap amino acid is in the terminal position. However, this structure has not been identified as a naturally occurring bacterial product. Our group and others 28,29 demonstrated that the naturally occurring peptidoglycan degradation product sensed by NOD1 is GlcNAc-MurNAc-L-Ala-D-Glumeso-DAP (GM-Tri DAP ). It is important to note that the recognition of these products by NOD1 is dependent on the presence of an exposed meso-dap, 25 an amino acid absent from eukaryotes, which, therefore, represents a very simple signature of bacterial origin. Meso-DAP is a characteristic feature of most Gram-negative bacteria and a few Gram-positive bacteria, such as Bacillus spp. Thus, the specificity of NOD1 to sense the presence of bacteria might represent a selective advantage for the host in certain cases when Gram-negative bacteria represent the main threat, such as in the epithelial cells lining the intestinal mucosa. In contrast, biochemical and functional analyses revealed that the muramyl dipeptide MurNAc-L-Ala-DisoGln (MDP) is the minimal motif sensed by NOD2. Interestingly, MDP is the minimal component common to the PG of both Gram-negative and Gram-positive bacteria, making this molecule a general sensor of bacterial infection through the recognition of this motif. However, MDP is not found as a naturally occurring bacterial product, but the corresponding muropeptide, GlcNAc- MurNAc-L-Ala-D-isoGln (GM-Di), does exist. Importantly, NOD2 has also been identified as the first susceptibility gene involved in the etiology of Crohn s disease, 30,31 an inflammatory bowel disease known to be influenced by both genetic and environmental factors. 32 The most common mutation in NOD2 associated with Crohn s disease (a frameshift mutation responsible for the truncation of the C-terminal LRR) results in a protein product that no longer detects peptidoglycan and MDP. 33,34 The implications of these findings remain poorly understood; however, it appears that defects in bacterial sensing may contribute to the etiology of this disease, at least in some cases. Indeed, a loss of surveillance activity by NOD2 may lead to defective local responses in the intestinal mucosa to bacterial infections, thereby

4 How Toll-like receptors and Nod-like receptors contribute to innate immunity in mammals 393 initiating systemic responses likely responsible for aberrant inflammation. Recently, two studies 35,36 have illustrated new functions of Nod2 in mice that might prove of great importance for the understanding of the implication of NOD2 mutations in the onset of Crohn s disease in humans. The first report demonstrates that Nod2 knock-out mice do not control intestinal infection with L. monocytogenes, and it is proposed that this defect results from an altered expression of cryptdins, 35 a class of defensins expressed in the intestine. 37 The second study describes the phenotype of a knock-in of a NOD2 frameshift mutation, which mimicks the most common NOD2 mutation occuring in the human gene. 36 Surprisingly, in this murine model, the frameshift mutation behaves as a gain-of-function modification; indeed, macrophages from the knock-in animals display constitutive activation of the NF-κB pathway and secrete mature IL-1β without stimulation. Even more unexpected is the fact that these cells are more potently stimulated by MDP than macrophages from wild-type animals. These results contradict the results obtained so far in the human system, using peripheral blood mononuclear cells from Nod2 frameshift patients. 38,39 Therefore, further studies will be needed to reconcile these different model systems. Only then will we be able to clarify the precise nature of the defect in NOD2 function responsible for the onset of Crohn s disease. Is there a crosstalk between TLR and NLR pathways? A great challenge for the coming years will be understanding how both the TLR and NLR pathways co-operate to mount an appropriate immune protection. Little is known, so far, about how these signaling pathways interconnect. However, future research will undoubtedly take advantage of several decades of work on muramyl peptides (and especially MDP), research which was performed before the discovery of NOD2. 40 For instance, a considerable body of evidence has demonstrated that MDP acts synergistically with LPS to induce inflammatory responses in myeloid cells. 41 The nature of such synergy is far from being understood, but several mechanisms have been proposed, including: (i) the up-regulation of MyD88 expression in MDP stimulated cells; 41 and (ii) the modulation of TLR signaling via an interaction between NOD2 and TAK1, 42 which is a downstream intermediate in TLR-mediated signaling pathways. Importantly, MDP-primed mice display enhanced sensitivity to LPS-induced septic shock and anaphylactic reactions. Recently, by monitoring, with precision, TNF secretion in human macrophages stimulated with varying concentrations of LPS and MDP, Hermann and co-workers 43 demonstrated that MDP could increase the sensitivity to LPS by up to three orders of magnitude. By using NOD1 and NOD2agonists, we recently provided further insights by demonstrating that both PG fragments synergize with TLR agonists to stimulate monocyte and DC proinflammatory cytokine production and DC maturation. 44 In naturally occurring pathological conditions such as septic shock or bacterial infection, PG fragments and LPS are likely to be found simultaneously in the same tissues, which further reinforces the notion that such synergy might be of great physiological importance. But is synergistic activation of cytokines by TLR and NOD agonists the only feature which links these two signaling pathways? Recently, Watanabe and co-workers 45 added a new and unexpected level of complexity to this picture; their work identified Nod2 as a negative regulator of TLR-2-mediated induction of IL-12 and IFN-γ in splenocytes. In contrast, another report found no difference in IL-12 production after stimulation of bone-marrow derived dendritic cells or macrophages from wild-type and Nod2 / mice upon TLR-2 stimulation. 35 These results suggest that any potential negative role of Nod2 in the TLR-2 response is not a universal phenomenon. Thus, it remains to be determined as to whether there is a contribution to TLR-2 responses from specific experimental procedures, genetic background, or cell types examined. Furthermore, future studies will be required to evaluate if these observations are relevant to the pathophysiology of Crohn s disease. Indeed, the study was conducted in mice invalidated for the Nod2 gene, while Crohn s disease is a human pathology in which the patients possess mutated forms of NOD2, which is not equivalent to a gene knockout. In line with this observation, it must be noted that Netea and coworkers 39 found that macrophages isolated from NOD2 frameshift mutation Crohn s disease patients display altered cytokine response to Pam 3, a synthetic agonist known to activate TLR-2 specifically. Future studies will allow a better understanding of how this crosstalk between the TLR and Nod pathways occurs. Until recently, it was an undisputed fact that TLR-2 could detect PG. However, recent evidence from our group has challenged this observation by showing that purification of PG from crude cell wall extracts to high purity results in progressive loss of TLR-2-dependent sensing. This strongly suggests that the motif activating TLR-2 in crude PG is due to contaminants that are present in most preparations (e.g. lipoproteins and LTA). Even though the exact nature of the contaminants responsible for TLR-2-dependent activity needs to be unambiguously characterized, these observations have the important consequence that Nod molecules and TLRs most probably do not detect overlapping microbial motifs. This is of great importance when considering the relative contribution of one PRM family versus the other in mammalian innate immunity.

5 394 Fritz, Girardin REFERENCES 1. Fearon DT, Locksley RM. The instructive role of innate immunity in the acquired immune response. Science 1996; 272: Girardin SE, Philpott DJ, Lemaitre B. Sensing microbes by diverse hosts. EMBO Report 2003; 4: Lemaitre B. The road to Toll. Nat Rev Immunol 2004; 4: Akira S. Mammalian Toll-like receptors. Curr Opin Immunol 2003; 15: Carneiro LA, Travassos LH, Philpott DJ. Innate immune recognition of microbes through Nod1 and Nod2: implications for disease. Microbes Infect 2004; 6: Akira S, Takeda K. Toll-like receptor signalling. Nat Rev Immunol 2004; 4: Lien E, Golenbock DT. Adjuvants and their signaling pathways: beyond TLRs. Nat Immunol 2003; 4: Hayashi F. Smith KD, Ozinsky et al. The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 2001; 410: Hemmi H, Takeuchi O, Kawai T et al. A Toll-like receptor recognizes bacterial DNA. Nature 2000; 408: Zhang D, Zhang MS, Hayden MS et al. A Toll-like receptor that prevents infection by uropathogenic bacteria. Science 2004; 303: Underhill DM. Toll-like receptors: networking for success. Eur J Immunol 2003; 33: Ozinsky A, Underhill JD, Fontenot JD et al. The repertoire for pattern recognition of pathogens by the innate immune system is defined by cooperation between Toll-like receptors. Proc Natl Acad Sci USA 2000; 97: Morr M, Takeuchi O, Akira S, Simon MM, Muhlradt PF et al. Differential recognition of structural details of bacterial lipopeptides by Toll-like receptors. Eur J Immunol 2002; 32: Takeuchi O, Sato S, Horiuchi T et al. Role of Toll-like receptor 1 in mediating immune response to microbial lipoproteins. J Immunol 2002; 169: Travassos LH, Girardin SE, Philpott DJ et al. Toll-like receptor 2-dependent bacterial sensing does not occur via peptidoglycan recognition. EMBO Report 2004; 5: Kaisho T, Akira S. Pleiotropic function of Toll-like receptors. Microbes Infect 2004; 6: Edelson BT, Unanue ER. MyD88-dependent but Toll-like receptor 2-independent innate immunity to Listeria: no role for either in macrophage listericidal activity. J Immunol 2002; 169: Haziot A, Hijiya N, Gangloff SC, Silver J, Goyert SM. Induction of a novel mechanism of accelerated bacterial clearance by lipopolysaccharide in CD14-deficient and Toll-like receptor 4- deficient mice. J Immunol 2001; 166: Latz E et al. TLR9 signals after translocating from the ER to CpG DNA in the lysosome. Nat Immunol 2004; 5: Hornef MW, Normark BH, Vandewalle A, Normark S. Intracellular recognition of lipopolysaccharide by Toll-like receptor 4 in intestinal epithelial cells. J Exp Med 2003; 198: Cossart P, Sansonetti PJ. Bacterial invasion: the paradigms of entero-invasive pathogens. Science 2004; 304: Hornef MW, Wick MJ, Rhen M, Normark S. Bacterial strategies for overcoming host innate and adaptive immune responses. Nat Immunol 2002; 3: Girardin SE, Tournebize R, Mavris M et al. CARD4/Nod1 mediates NF-kappaB and JNK activation by invasive Shigella flexneri. EMBO Report 2001; 2: Ting JP, Davis BK. CATERPILLER: a novel gene family important in immunity, cell death, and diseases. Annu Rev Immunol 2005; 23: Girardin SE, Travassos LH, Herve M et al. Peptidoglycan molecular requirements allowing detection by Nod1 and Nod2. J Biol Chem Girardin SE, Philpott DJ. The role of peptidoglycan recognition in innate immunity. Eur J Immunol 2004; 34: Boneca IG. The role of peptidoglycan in pathogenesis. Curr Opin Microbiol 2005; 8: Girardin SE, Boneca IG, Carneiro LA et al. Nod1 detects a unique muropeptide from Gram-negative bacterial peptidoglycan. Science 2003; 300: Chamaillard M, Hashimoto, Horie Y et al. An essential role for NOD1 in host recognition of bacterial peptidoglycan containing diaminopimelic acid. Nat Immunol 2003; 4: Hugot JP, Chamaillard M, Zouali H et al. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn s disease. Nature 2001; 411: Ogura Y, Bonen DK, Inohara N et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn s disease. Nature 2001; 411: Bouma G, Strober W. The immunological and genetic basis of inflammatory bowel disease. Nat Rev Immunol 2003; 3: Chamaillard M, Philpott D, Girardin SE et al. Gene-environment interaction modulated by allelic heterogeneity in inflammatory diseases. Proc Natl Acad Sci USA 2003; 100: Ogura Y, Saab L, Chen FF et al. Genetic variation and activity of mouse Nod2, a susceptibility gene for Crohn s disease small star, filled. Genomics 2003; 81: Kobayashi KS, Chamaillard M, Ogura Y et al. Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science 2005; 307: Maeda S, Hsu LC, Liu H et al. Nod2 mutation in Crohn s disease potentiates NF-kappaB activity and IL-1beta processing. Science 2005; 307: Ayabe T, Ashida T, Kohgo Y, Kono T. The role of Paneth cells and their antimicrobial peptides in innate host defense. Trends Microbiol 2004; 12: Li J, Moran T, Swanson E et al. Regulation of IL-8 and IL-1β expression in Crohn s disease associated NOD2/CARD15 mutations. Hum Mol Genet Netea MG, Kullberg BJ, De Jong DJ et al. NOD2 mediates antiinflammatory signals induced by TLR2 ligands: implications for Crohn s disease. Eur J Immunol 2004; 34: Adam A, Petit JF, Lrfrancier P, Lederer E. Muramyl peptides. Chemical structure, biological activity and mechanism of action. Mol Cell Biochem 1981; 41: Takada H, Yokoyama S, Yang S. Enhancement of endotoxin activity by muramyldipeptide. J Endotoxin Res 2002; 8: Chen CM et al. Reciprocal cross-talk between Nod2 and TAK1 signaling pathways. J Biol Chem 2004; 279: Traub S, Kubasch N, Morath S et al. Structural requirements of synthetic muropeptides to synergize with lipopolysaccharide in cytokine induction. J Biol Chem 2004; 279: Fritz JH, Girardin SE, Fitting C et al. Synergistic stimulation of human monocytes and dendritic cells by Toll-like receptor 4 and NOD1- and NOD2- activating agonists. Eur J Immunol 2005; 35: Watanabe T, Kitani A, Murray PJ, Strober W. NOD2 is a negative regulator of Toll-like receptor 2-mediated T helper type 1 responses. Nat Immunol 2004; 5: