- Email: [email protected]
Innate immunity: Lipoproteins take their Toll on the host
Dispatch
R879
Innate immunity: Lipoproteins take their Toll on the host Charles A. Janeway, Jr and Ruslan Medzhitov
A family of mammalian Toll-like receptors (TLRs) has a critical role in the recognition of microbial infection. Recent evidence suggests that bacterial lipoproteins — major components of bacterial cell walls — are recognized by a member of the human TLR family. Address: Yale University School of Medicine, Section of Immunobiology, 310 Cedar Street, LH416, P.O. Box 208011, New Haven, Connecticut 06520-8011, USA. Current Biology 1999, 9:R879–R882 0960-9822/99/$ – see front matter © 1999 Elsevier Science Ltd. All rights reserved.
The recognition of infecting microbes followed by the induction of an effective immune response is essential for the survival of most multicellular organisms. It is equally important, however, that the immune response is not induced upon the recognition of self antigens, or non-infectious, innocuous, non-self antigens. So, how does the immune system decide which antigens to respond to and which to ignore? We believe that the immune system must somehow be able to determine the origin of the antigens, such that only the antigens of microbial origin would induce immune responses, whereas self antigens and non-microbial, non-self antigens would not. The antigens themselves, however, have no special features that are intrinsically associated with their origin and, indeed, conventional T and B lymphocytes are not preprogrammed to recognize just microbial antigens. It has been suggested [1] that host organisms detect the presence of infection by recognizing a limited number of conserved structures produced only by micro-organisms and not by multicellular hosts. These structures are referred to as pathogen-associated molecular patterns (PAMPs) and their recognition is followed by the induction of inflammatory responses required for the recruitment and activation of lymphocytes. Receptors that recognize these molecular patterns have been an important missing link in connecting the detection of infection with the activation of antigen-specific lymphocytes. A family of mammalian receptors, called Toll-like receptors (TLRs), has recently been identified [2,3] that appears to perform just this function. The identification of TLRs prompted many researchers to test their favorite PAMPs for recognition by individual members of the Toll family. Three recent reports [4–6] provide evidence that bacterial lipoproteins — one of the major PAMPs produced by all bacteria — are recognized by a particular member of the TLR family, TLR2.
Mammalian TLRs, as the name implies, are homologs of the Drosophila Toll protein, which was originally identified as a receptor involved in the control of dorso-ventral pattern formation in fly embryos [7]. Both Drosophila and mammalian Toll receptors are transmembrane proteins with a large extracellular domain that contains multiple leucine-rich repeats. The receptors also contain a cytoplasmic domain homologous to that of the interleukin-1 (IL-1) receptor, and therefore referred to as a TIR domain, for Toll/IL-1 receptor homology domain (Figure 1). The similarities between members of the Toll family and members of the IL-1 receptor family are not restricted to their structure, however, because both families induce the activation of the transcription factor NF-κB and share many components of the NF-κB signaling pathway [8]. The first evidence of the involvement of Toll signaling in host defense came from the analysis of immune responses in Drosophila mutants carrying loss-of-function mutations in various components of the Toll pathway [9]. As it turned out, flies deficient in several individual components of the Toll signaling pathway were unable to recognize fungal infection and produce drosomycin, a major antifungal peptide [9]. More recently, another member of the Drosophila Toll family, 18-Wheeler, has been demonstrated to be involved in the recognition of bacterial infection [10]. Microbial infection in Drosophila and other insects results in a rapid induction of a battery of antimicrobial peptides that kill pathogens by punching holes in their cell walls [11]. Importantly, individual peptides have selective activity against a particular class of micro-organisms — for example, gram-positive or gram-negative bacteria, or fungi — and infection with a given class of pathogens results in the preferential induction of only the appropriate peptides [12]. As the induction of the different antimicrobial peptides is likely to be controlled by individual members of the Toll family, it appears that in Drosophila the Toll receptors not only detect the presence of infection, but also discriminate between different classes of pathogens. The recent studies of the human and mouse TLRs suggest an intriguing possibility that the ability of Toll receptors to recognize a particular pathogen class may have been conserved in mammals. Indeed, mouse TLR4 has been shown to be a receptor for lipopolysaccharide — a major PAMP associated with gram-negative bacteria [13–15]. Moreover, a mouse strain that carries a spontaneous loss-of-function mutation in the tlr4 gene demonstrates selective susceptibility to gram-negative bacterial infections [13,14].
R880
Current Biology Vol 9 No 23
Figure 1 Gram-negative Gram-positive Mycobacteria bacteria bacteria LPS, BLP
LTA, BLP
LAM, BLP
Fungi
Pathogens
Mannans PAMPs
TLR Antigen-presenting cell
Expression of cytokines, co-stimulators and chemokines
Inflammation, induction of adaptive immune responses Current Biology
TLRs recognize pathogen-associated molecular patterns (PAMPS) that often represent molecular signatures of a particular pathogen class: LPS (lipopolysaccharide) is a product of gram-negative bacteria; BLP (bacterial lipoproteins) are found in all bacteria; LTA (lipoteichoic acid) is a product of gram-positive bacteria; peptidoglycan is present in both gram-positive and gram-negative bacteria, but is exposed for recognition only in gram-positive bacteria; LAM (lipoarabinomannan) is a product of mycobacteria; and mannans and glucans are products of yeast cell walls. Recognition of these various molecules by TLRs on antigen-presenting cells activates the expression of various factors that lead to inflammatory responses and the induction of adaptive immunity.
Additionally, biochemical data suggest that another member of the Toll family, TLR2, is involved in the recognition of gram-positive bacteria, probably by recognizing peptidoglycan, a PAMP found in all bacteria, but only exposed for recognition in the gram-positive microbes [16,17]. The most compelling evidence for a differential role of TLRs in pathogen class recognition comes from the comparative analysis of TLR2- and TLR4-deficient mice [18]. A detailed characterization of lipopolysaccharide responsiveness in these mice clearly demonstrates that, at least in the mouse, TLR2 is not necessary for the lipopolysaccharide responses. Indeed, the susceptibility to endotoxic shock, cytokine production, and the activation of signaling pathways induced by lipopolysaccharide are normal in the TLR2 knockout mice, whereas all these responses are completely eliminated in the TLR4-deficient mice [18]. Conversely, in the TLR2 knockout mice, the cellular responses to gram-positive bacterial cell walls are severely impaired, and the responses to peptidoglycan are eliminated completely, whereas these responses are normal in the TLR4
knockout mice [18]. These results indicate that TLR2 is involved in the recognition of peptidoglycan and gram-positive bacteria, and TLR4 is specialized for the recognition of lipopolysaccharide and, therefore, gram-negative bacteria. Surprisingly, however, cells derived from mice deficient in TLR4, but not TLR2, are also unresponsive to lipoteichoic acid — a PAMP found only in gram-positive bacteria [18]. This is a very important finding that suggests that different TLRs may have evolved to recognize PAMPs that have similar structural patterns rather than a common microbial origin. Indeed, the molecular pattern of lipoteichoic acid is quite similar to lipopolysaccharide and completely distinct from peptidoglycan. It is perhaps premature to make this generalization, however, given the recent publications that add bacterial lipoproteins to the growing number of PAMPs recognized by mammalian TLRs [4–6]. These lipoproteins are found in all bacteria and share a common chemical motif — covalently bound, highly hydrophobic acyl groups. Interestingly, as with lipopolysaccharide, the lipid moiety of bacterial lipoproteins is necessary for their function in bacteria and for their recognition by the host’s innate immune system [19]. This reflects the strategy of the innate immune system to recognize molecular patterns essential for the survival of the micro-organisms, thus preventing the generation of escape mutants that would therefore lack the ability to cause infection. Several types of experiments have revealed that bacterial lipoproteins signal through human TLR2: first, expression of TLR2 in TLR2-negative cell lines confers responsiveness to bacterial lipoproteins; second, a TLR2-specific monoclonal antibody blocks bacterial lipoprotein-induced cell signaling; and third, expression of a dominant-negative inhibitory form of TLR2 blocks the response to bacterial lipoproteins [4–6]. Finally, and most convincingly, responses induced by bacterial lipoproteins are completely eliminated in cells derived from the TLR2 knockout mice (S. Akira, personal communication). There are two key implications of these findings. Unlike lipopolysaccharide and lipoteichoic acid, bacterial lipoproteins are not restricted to any particular type of bacteria, and TLR2mediated recognition of bacterial lipoproteins may therefore not discriminate between different classes of bacteria. Also, bacterial lipoproteins do not share any obvious structural patterns with peptidoglycans, suggesting that at least some TLRs can be involved in the recognition of chemically unrelated PAMPs. The recent data implicating TLR2 in the recognition of zymosan [20], a component of yeast cell walls, lend support to both of these points: on the one hand, zymosan is structurally unrelated to both peptidoglycan and bacterial lipoproteins; on the other hand, given that it is a yeast molecule, zymosan does not share a common microbial origin with peptidoglycan and bacterial lipoproteins, which are present only in bacteria.
Dispatch
At this stage, it is evident that the role of TLR2 in the recognition of infection is far from being completely resolved. The demonstration that TLR2 knockout mice do not have any obvious defects in responses to lipopolysaccharide and lipoteichoic acid [18] is clearly at odds with previously published reports implicating human TLR2 in the recognition of these molecules [17,21,22]. There is also something unsettling in the fact that the same receptor (TLR2) has been found to be involved in the recognition of lipopolysaccharide [21–24], lipoteichoic acid, peptidoglycan [16,17], bacterial lipoproteins [4–6], and lipoarabinomannan [25] in studies performed in vitro. At the same time, TLR2 does not seem to be necessary for lipopolysaccharide recognition in a hamster cell line [26], whereas both mouse and human TLR4 seem to mediate lipopolysaccharide signaling [13–15,23,24]. It is unclear at present whether these discrepancies reflect species-specific differences in TLR recognition, or perhaps some subtle flaws in the experimental design of the in vitro studies. The latter may be due to any one of a number of factors. Overexpression of TLRs may have some non-specific effect on the responsiveness of the reporter cell lines. Also, most (but not all) reports lack some essential control experiments: for example, when expression of a wild-type or a dominant-negative version of TLR2 is used to confer or inhibit responsiveness to a particular stimulus, a proper control would be the use of wild-type or dominant-negative forms of TLR other than TLR2. The response monitored in most reports is the activation of NF-κB, which may not always reflect the physiological cellular responses mediated by TLRs: for example, lipopolysaccharide can still activate NF-κB in mice that lack the Toll/IL-1 signaling adaptor molecule MyD88 and are defective in TLR2 and TLR4 signaling. The cellular responses, however, as measured by cytokine production and the induction of septic shock, are completely eliminated in these animals [27]. The reason for the lack of correlation between NF-κB activation and the induction of NF-κB-dependent inflammatory cytokines in these mice is currently unknown, but possibly suggests that NF-κB activation, although necessary, may not be an adequate ‘read-out’ for TLR activation. Finally, it should be emphasized that at present we do not really know the identity of the actual ligands for any of the TLRs. It may well be that, as with Drosophila Toll, mammalian TLRs do not recognize PAMPs directly, but lie on a pathway downstream of the actual pattern recognition event. If this is the case, the gene products involved in the generation of the ligands may be differentially expressed in the various cell lines used for the in vitro studies, thus accounting for the observed discrepancies in the TLR specificities.
R881
Thus, the field of the TLRs is in its infancy. It will be necessary to generate mice deficient in several more TLRs to study their specific and shared ligands. Upstream and downstream modifiers of pathogen responses also need to be characterized. It may be that it is not simply one TLR that tells the adaptive immune system what kind of pathogen is on the premises, but that some kind of integrated signal informs the host adaptive immune system what kind of response is needed to combat the particular pathogen. As an example, we found that, although both TLR4 and IL-1 can induce NF-κB, Toll but not IL-1 activates the AP-1 transcription factor (our unpublished observations). The integration of signals from one or more TLRs seems more likely to be the driving force in determining the outcome of the response, and this means that extensive studies of mice lacking various TLRs and of the effects of these deficiencies on various infectious agents need to be carried out. This is a tall order, but it is likely to yield decisive information on the biological function of each TLR. References 1. Janeway CA, Jr: Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harb Symp Quant Biol 1989, 54:1-13. 2. Medzhitov R, Preston-Hurlburt P, Janeway CA, Jr: A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 1997, 388:394-397. 3. Rock FL, Hardiman G, Timans JC, Kastelein RA, Bazan JF: A family of human receptors structurally related to Drosophila Toll. Proc Natl Acad Sci USA 1998, 95:588-593. 4. Brightbill HD, Libraty DH, Krutzik SR, Yang RB, Belisle JT, Bleharski JR, Maitland M, Norgard MV, Plevy SE, Smale ST, et al.: Host defense mechanisms triggered by microbial lipoproteins through toll-like receptors. Science 1999, 285:732-736. 5. Aliprantis AO, Yang RB, Mark MR, Suggett S, Devaux B, Radolf JD, Klimpel GR, Godowski P, Zychlinsky A: Cell activation and apoptosis by bacterial lipoproteins through toll-like receptor-2. Science 1999, 285:736-739. 6. Hirschfeld M, Kirschning CJ, Schwandner R, Wesche H, Weis JH, Wooten RM, Weis JJ: Cutting edge: inflammatory signaling by Borrelia burgdorferi lipoproteins is mediated by toll-like receptor 2. J Immunol 1999,163:2382-2386. 7. Hashimoto C, Hudson KL, Anderson KV: The Toll gene of Drosophila, required for dorsal-ventral embryonic polarity, appears to encode a transmembrane protein. Cell 1988, 52:269-279. 8. Kopp EB, Medzhitov R: The Toll-receptor family and control of innate immunity. Curr Opin Immunol 1999, 11:13-18. 9. Lemaitre B, Nicolas E, Michaut L, Reichhart JM, Hoffmann JA: The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 1996, 86:973-983. 10. Williams MJ, Rodriguez A, Kimbrell DA, Eldon ED: The 18-wheeler mutation reveals complex antibacterial gene regulation in Drosophila host defense. EMBO J 1997, 16:6120-6130. 11. Hoffmann JA, Kafatos FC, Janeway CA, Ezekowitz RA: Phylogenetic perspectives in innate immunity. Science 1999, 284:1313-1318. 12. Lemaitre B, Reichhart JM, Hoffmann JA: Drosophila host defense: differential induction of antimicrobial peptide genes after infection by various classes of microorganisms. Proc Natl Acad Sci USA 1997, 94:14614-14619. 13. Poltorak A, He X, Smirnova I, Liu MY, Huffel CV, Du X, Birdwell D, Alejos E, Silva M, Galanos C, et al.: Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 1998, 282:2085-2088. 14. Qureshi ST, Lariviere L, Leveque G, Clermont S, Moore KJ, Gros P, Malo D: Endotoxin-tolerant mice have mutations in Toll-like receptor 4 [Tlr4]. J Exp Med 1999, 189:615-625.
R882
Current Biology Vol 9 No 23
15. Hoshino K, Takeuchi O, Kawai T, Sanjo H, Ogawa T, Takeda Y, Takeda K, Akira S: Cutting edge: Toll-like receptor 4 [TLR4]deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J Immunol 1999, 162:3749-3752. 16. Yoshimura A, Lien E, Ingalls RR, Tuomanen E, Dziarski R, Golenbock D: Cutting edge: recognition of Gram-positive bacterial cell wall components by the innate immune system occurs via Toll-like receptor 2. J Immunol 1999, 163:1-5. 17. Schwandner R, Dziarski R, Wesche H, Rothe M, Kirschning CJ: Peptidoglycan- and lipoteichoic acid-induced cell activation is mediated by toll-like receptor 2. J Biol Chem 1999, 274:17406-17409. 18. Takeuchi O, Hoshino K, Kawai T, Sanjo H, Takada H, Ogawa T, Takeda K, Akira S: Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram-positive bacterial cell wall components. Immunity 1999, 11:443-451. 19. Bessler WG, Cox M, Lex A, Suhr B, Wiesmuller KH, Jung G: Synthetic lipopeptide analogs of bacterial lipoprotein are potent polyclonal activators for murine B lymphocytes. J Immunol 1985, 135:1900-1905. 20. Underhill D, Ozinsky A, Hajjar A, Stevens A, Wilson C, Bassetti M, Aderem A: The Toll-like receptor 2 is recruited to macrophage phagosomes and discriminates between pathogens. Nature 1999, 401:811-815. 21. Yang RB, Mark MR, Gray A, Huang A, Xie MH, Zhang M, Goddard A, Wood WI, Gurney AL, Godowski PJ: Toll-like receptor-2 mediates lipopolysaccharide-induced cellular signalling. Nature 1998, 395:284-288. 22. Kirschning CJ, Wesche H, Merrill Ayres T, Rothe M: Human toll-like receptor 2 confers responsiveness to bacterial lipopolysaccharide. J Exp Med 1998, 188:2091-2097. 23. Shimazu R, Akashi S, Ogata H, Nagai Y, Fukudome K, Miyake K, Kimoto M: MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4. J Exp Med 1999,189:1777-1782. 24. Chow JC, Young DW, Golenbock DT, Christ WJ, Gusovsky F: Tolllike receptor-4 mediates lipopolysaccharide-induced signal transduction. J Biol Chem 1999, 274:10689-10692. 25. Means TK, Wang S, Lien E, Yoshimura A, Golenbock DT, Fenton MJ: Human toll-like receptors mediate cellular activation by Mycobacterium tuberculosis. J Immunol 1999, 163:3920-3927. 26. Heine H, Kirschning CJ, Lien E, Monks BG, Rothe M, Golenbock DT: Cutting edge: cells that carry a null allele for toll-like receptor 2 are capable of responding to endotoxin. J Immunol 1999, 162:6971-6975. 27. Kawai T, Adachi O, Ogawa T, Takeda K, Akira S: Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity 1999, 11:115-122.
R879
Innate immunity: Lipoproteins take their Toll on the host Charles A. Janeway, Jr and Ruslan Medzhitov
A family of mammalian Toll-like receptors (TLRs) has a critical role in the recognition of microbial infection. Recent evidence suggests that bacterial lipoproteins — major components of bacterial cell walls — are recognized by a member of the human TLR family. Address: Yale University School of Medicine, Section of Immunobiology, 310 Cedar Street, LH416, P.O. Box 208011, New Haven, Connecticut 06520-8011, USA. Current Biology 1999, 9:R879–R882 0960-9822/99/$ – see front matter © 1999 Elsevier Science Ltd. All rights reserved.
The recognition of infecting microbes followed by the induction of an effective immune response is essential for the survival of most multicellular organisms. It is equally important, however, that the immune response is not induced upon the recognition of self antigens, or non-infectious, innocuous, non-self antigens. So, how does the immune system decide which antigens to respond to and which to ignore? We believe that the immune system must somehow be able to determine the origin of the antigens, such that only the antigens of microbial origin would induce immune responses, whereas self antigens and non-microbial, non-self antigens would not. The antigens themselves, however, have no special features that are intrinsically associated with their origin and, indeed, conventional T and B lymphocytes are not preprogrammed to recognize just microbial antigens. It has been suggested [1] that host organisms detect the presence of infection by recognizing a limited number of conserved structures produced only by micro-organisms and not by multicellular hosts. These structures are referred to as pathogen-associated molecular patterns (PAMPs) and their recognition is followed by the induction of inflammatory responses required for the recruitment and activation of lymphocytes. Receptors that recognize these molecular patterns have been an important missing link in connecting the detection of infection with the activation of antigen-specific lymphocytes. A family of mammalian receptors, called Toll-like receptors (TLRs), has recently been identified [2,3] that appears to perform just this function. The identification of TLRs prompted many researchers to test their favorite PAMPs for recognition by individual members of the Toll family. Three recent reports [4–6] provide evidence that bacterial lipoproteins — one of the major PAMPs produced by all bacteria — are recognized by a particular member of the TLR family, TLR2.
Mammalian TLRs, as the name implies, are homologs of the Drosophila Toll protein, which was originally identified as a receptor involved in the control of dorso-ventral pattern formation in fly embryos [7]. Both Drosophila and mammalian Toll receptors are transmembrane proteins with a large extracellular domain that contains multiple leucine-rich repeats. The receptors also contain a cytoplasmic domain homologous to that of the interleukin-1 (IL-1) receptor, and therefore referred to as a TIR domain, for Toll/IL-1 receptor homology domain (Figure 1). The similarities between members of the Toll family and members of the IL-1 receptor family are not restricted to their structure, however, because both families induce the activation of the transcription factor NF-κB and share many components of the NF-κB signaling pathway [8]. The first evidence of the involvement of Toll signaling in host defense came from the analysis of immune responses in Drosophila mutants carrying loss-of-function mutations in various components of the Toll pathway [9]. As it turned out, flies deficient in several individual components of the Toll signaling pathway were unable to recognize fungal infection and produce drosomycin, a major antifungal peptide [9]. More recently, another member of the Drosophila Toll family, 18-Wheeler, has been demonstrated to be involved in the recognition of bacterial infection [10]. Microbial infection in Drosophila and other insects results in a rapid induction of a battery of antimicrobial peptides that kill pathogens by punching holes in their cell walls [11]. Importantly, individual peptides have selective activity against a particular class of micro-organisms — for example, gram-positive or gram-negative bacteria, or fungi — and infection with a given class of pathogens results in the preferential induction of only the appropriate peptides [12]. As the induction of the different antimicrobial peptides is likely to be controlled by individual members of the Toll family, it appears that in Drosophila the Toll receptors not only detect the presence of infection, but also discriminate between different classes of pathogens. The recent studies of the human and mouse TLRs suggest an intriguing possibility that the ability of Toll receptors to recognize a particular pathogen class may have been conserved in mammals. Indeed, mouse TLR4 has been shown to be a receptor for lipopolysaccharide — a major PAMP associated with gram-negative bacteria [13–15]. Moreover, a mouse strain that carries a spontaneous loss-of-function mutation in the tlr4 gene demonstrates selective susceptibility to gram-negative bacterial infections [13,14].
R880
Current Biology Vol 9 No 23
Figure 1 Gram-negative Gram-positive Mycobacteria bacteria bacteria LPS, BLP
LTA, BLP
LAM, BLP
Fungi
Pathogens
Mannans PAMPs
TLR Antigen-presenting cell
Expression of cytokines, co-stimulators and chemokines
Inflammation, induction of adaptive immune responses Current Biology
TLRs recognize pathogen-associated molecular patterns (PAMPS) that often represent molecular signatures of a particular pathogen class: LPS (lipopolysaccharide) is a product of gram-negative bacteria; BLP (bacterial lipoproteins) are found in all bacteria; LTA (lipoteichoic acid) is a product of gram-positive bacteria; peptidoglycan is present in both gram-positive and gram-negative bacteria, but is exposed for recognition only in gram-positive bacteria; LAM (lipoarabinomannan) is a product of mycobacteria; and mannans and glucans are products of yeast cell walls. Recognition of these various molecules by TLRs on antigen-presenting cells activates the expression of various factors that lead to inflammatory responses and the induction of adaptive immunity.
Additionally, biochemical data suggest that another member of the Toll family, TLR2, is involved in the recognition of gram-positive bacteria, probably by recognizing peptidoglycan, a PAMP found in all bacteria, but only exposed for recognition in the gram-positive microbes [16,17]. The most compelling evidence for a differential role of TLRs in pathogen class recognition comes from the comparative analysis of TLR2- and TLR4-deficient mice [18]. A detailed characterization of lipopolysaccharide responsiveness in these mice clearly demonstrates that, at least in the mouse, TLR2 is not necessary for the lipopolysaccharide responses. Indeed, the susceptibility to endotoxic shock, cytokine production, and the activation of signaling pathways induced by lipopolysaccharide are normal in the TLR2 knockout mice, whereas all these responses are completely eliminated in the TLR4-deficient mice [18]. Conversely, in the TLR2 knockout mice, the cellular responses to gram-positive bacterial cell walls are severely impaired, and the responses to peptidoglycan are eliminated completely, whereas these responses are normal in the TLR4
knockout mice [18]. These results indicate that TLR2 is involved in the recognition of peptidoglycan and gram-positive bacteria, and TLR4 is specialized for the recognition of lipopolysaccharide and, therefore, gram-negative bacteria. Surprisingly, however, cells derived from mice deficient in TLR4, but not TLR2, are also unresponsive to lipoteichoic acid — a PAMP found only in gram-positive bacteria [18]. This is a very important finding that suggests that different TLRs may have evolved to recognize PAMPs that have similar structural patterns rather than a common microbial origin. Indeed, the molecular pattern of lipoteichoic acid is quite similar to lipopolysaccharide and completely distinct from peptidoglycan. It is perhaps premature to make this generalization, however, given the recent publications that add bacterial lipoproteins to the growing number of PAMPs recognized by mammalian TLRs [4–6]. These lipoproteins are found in all bacteria and share a common chemical motif — covalently bound, highly hydrophobic acyl groups. Interestingly, as with lipopolysaccharide, the lipid moiety of bacterial lipoproteins is necessary for their function in bacteria and for their recognition by the host’s innate immune system [19]. This reflects the strategy of the innate immune system to recognize molecular patterns essential for the survival of the micro-organisms, thus preventing the generation of escape mutants that would therefore lack the ability to cause infection. Several types of experiments have revealed that bacterial lipoproteins signal through human TLR2: first, expression of TLR2 in TLR2-negative cell lines confers responsiveness to bacterial lipoproteins; second, a TLR2-specific monoclonal antibody blocks bacterial lipoprotein-induced cell signaling; and third, expression of a dominant-negative inhibitory form of TLR2 blocks the response to bacterial lipoproteins [4–6]. Finally, and most convincingly, responses induced by bacterial lipoproteins are completely eliminated in cells derived from the TLR2 knockout mice (S. Akira, personal communication). There are two key implications of these findings. Unlike lipopolysaccharide and lipoteichoic acid, bacterial lipoproteins are not restricted to any particular type of bacteria, and TLR2mediated recognition of bacterial lipoproteins may therefore not discriminate between different classes of bacteria. Also, bacterial lipoproteins do not share any obvious structural patterns with peptidoglycans, suggesting that at least some TLRs can be involved in the recognition of chemically unrelated PAMPs. The recent data implicating TLR2 in the recognition of zymosan [20], a component of yeast cell walls, lend support to both of these points: on the one hand, zymosan is structurally unrelated to both peptidoglycan and bacterial lipoproteins; on the other hand, given that it is a yeast molecule, zymosan does not share a common microbial origin with peptidoglycan and bacterial lipoproteins, which are present only in bacteria.
Dispatch
At this stage, it is evident that the role of TLR2 in the recognition of infection is far from being completely resolved. The demonstration that TLR2 knockout mice do not have any obvious defects in responses to lipopolysaccharide and lipoteichoic acid [18] is clearly at odds with previously published reports implicating human TLR2 in the recognition of these molecules [17,21,22]. There is also something unsettling in the fact that the same receptor (TLR2) has been found to be involved in the recognition of lipopolysaccharide [21–24], lipoteichoic acid, peptidoglycan [16,17], bacterial lipoproteins [4–6], and lipoarabinomannan [25] in studies performed in vitro. At the same time, TLR2 does not seem to be necessary for lipopolysaccharide recognition in a hamster cell line [26], whereas both mouse and human TLR4 seem to mediate lipopolysaccharide signaling [13–15,23,24]. It is unclear at present whether these discrepancies reflect species-specific differences in TLR recognition, or perhaps some subtle flaws in the experimental design of the in vitro studies. The latter may be due to any one of a number of factors. Overexpression of TLRs may have some non-specific effect on the responsiveness of the reporter cell lines. Also, most (but not all) reports lack some essential control experiments: for example, when expression of a wild-type or a dominant-negative version of TLR2 is used to confer or inhibit responsiveness to a particular stimulus, a proper control would be the use of wild-type or dominant-negative forms of TLR other than TLR2. The response monitored in most reports is the activation of NF-κB, which may not always reflect the physiological cellular responses mediated by TLRs: for example, lipopolysaccharide can still activate NF-κB in mice that lack the Toll/IL-1 signaling adaptor molecule MyD88 and are defective in TLR2 and TLR4 signaling. The cellular responses, however, as measured by cytokine production and the induction of septic shock, are completely eliminated in these animals [27]. The reason for the lack of correlation between NF-κB activation and the induction of NF-κB-dependent inflammatory cytokines in these mice is currently unknown, but possibly suggests that NF-κB activation, although necessary, may not be an adequate ‘read-out’ for TLR activation. Finally, it should be emphasized that at present we do not really know the identity of the actual ligands for any of the TLRs. It may well be that, as with Drosophila Toll, mammalian TLRs do not recognize PAMPs directly, but lie on a pathway downstream of the actual pattern recognition event. If this is the case, the gene products involved in the generation of the ligands may be differentially expressed in the various cell lines used for the in vitro studies, thus accounting for the observed discrepancies in the TLR specificities.
R881
Thus, the field of the TLRs is in its infancy. It will be necessary to generate mice deficient in several more TLRs to study their specific and shared ligands. Upstream and downstream modifiers of pathogen responses also need to be characterized. It may be that it is not simply one TLR that tells the adaptive immune system what kind of pathogen is on the premises, but that some kind of integrated signal informs the host adaptive immune system what kind of response is needed to combat the particular pathogen. As an example, we found that, although both TLR4 and IL-1 can induce NF-κB, Toll but not IL-1 activates the AP-1 transcription factor (our unpublished observations). The integration of signals from one or more TLRs seems more likely to be the driving force in determining the outcome of the response, and this means that extensive studies of mice lacking various TLRs and of the effects of these deficiencies on various infectious agents need to be carried out. This is a tall order, but it is likely to yield decisive information on the biological function of each TLR. References 1. Janeway CA, Jr: Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harb Symp Quant Biol 1989, 54:1-13. 2. Medzhitov R, Preston-Hurlburt P, Janeway CA, Jr: A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 1997, 388:394-397. 3. Rock FL, Hardiman G, Timans JC, Kastelein RA, Bazan JF: A family of human receptors structurally related to Drosophila Toll. Proc Natl Acad Sci USA 1998, 95:588-593. 4. Brightbill HD, Libraty DH, Krutzik SR, Yang RB, Belisle JT, Bleharski JR, Maitland M, Norgard MV, Plevy SE, Smale ST, et al.: Host defense mechanisms triggered by microbial lipoproteins through toll-like receptors. Science 1999, 285:732-736. 5. Aliprantis AO, Yang RB, Mark MR, Suggett S, Devaux B, Radolf JD, Klimpel GR, Godowski P, Zychlinsky A: Cell activation and apoptosis by bacterial lipoproteins through toll-like receptor-2. Science 1999, 285:736-739. 6. Hirschfeld M, Kirschning CJ, Schwandner R, Wesche H, Weis JH, Wooten RM, Weis JJ: Cutting edge: inflammatory signaling by Borrelia burgdorferi lipoproteins is mediated by toll-like receptor 2. J Immunol 1999,163:2382-2386. 7. Hashimoto C, Hudson KL, Anderson KV: The Toll gene of Drosophila, required for dorsal-ventral embryonic polarity, appears to encode a transmembrane protein. Cell 1988, 52:269-279. 8. Kopp EB, Medzhitov R: The Toll-receptor family and control of innate immunity. Curr Opin Immunol 1999, 11:13-18. 9. Lemaitre B, Nicolas E, Michaut L, Reichhart JM, Hoffmann JA: The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 1996, 86:973-983. 10. Williams MJ, Rodriguez A, Kimbrell DA, Eldon ED: The 18-wheeler mutation reveals complex antibacterial gene regulation in Drosophila host defense. EMBO J 1997, 16:6120-6130. 11. Hoffmann JA, Kafatos FC, Janeway CA, Ezekowitz RA: Phylogenetic perspectives in innate immunity. Science 1999, 284:1313-1318. 12. Lemaitre B, Reichhart JM, Hoffmann JA: Drosophila host defense: differential induction of antimicrobial peptide genes after infection by various classes of microorganisms. Proc Natl Acad Sci USA 1997, 94:14614-14619. 13. Poltorak A, He X, Smirnova I, Liu MY, Huffel CV, Du X, Birdwell D, Alejos E, Silva M, Galanos C, et al.: Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 1998, 282:2085-2088. 14. Qureshi ST, Lariviere L, Leveque G, Clermont S, Moore KJ, Gros P, Malo D: Endotoxin-tolerant mice have mutations in Toll-like receptor 4 [Tlr4]. J Exp Med 1999, 189:615-625.
R882
Current Biology Vol 9 No 23
15. Hoshino K, Takeuchi O, Kawai T, Sanjo H, Ogawa T, Takeda Y, Takeda K, Akira S: Cutting edge: Toll-like receptor 4 [TLR4]deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J Immunol 1999, 162:3749-3752. 16. Yoshimura A, Lien E, Ingalls RR, Tuomanen E, Dziarski R, Golenbock D: Cutting edge: recognition of Gram-positive bacterial cell wall components by the innate immune system occurs via Toll-like receptor 2. J Immunol 1999, 163:1-5. 17. Schwandner R, Dziarski R, Wesche H, Rothe M, Kirschning CJ: Peptidoglycan- and lipoteichoic acid-induced cell activation is mediated by toll-like receptor 2. J Biol Chem 1999, 274:17406-17409. 18. Takeuchi O, Hoshino K, Kawai T, Sanjo H, Takada H, Ogawa T, Takeda K, Akira S: Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram-positive bacterial cell wall components. Immunity 1999, 11:443-451. 19. Bessler WG, Cox M, Lex A, Suhr B, Wiesmuller KH, Jung G: Synthetic lipopeptide analogs of bacterial lipoprotein are potent polyclonal activators for murine B lymphocytes. J Immunol 1985, 135:1900-1905. 20. Underhill D, Ozinsky A, Hajjar A, Stevens A, Wilson C, Bassetti M, Aderem A: The Toll-like receptor 2 is recruited to macrophage phagosomes and discriminates between pathogens. Nature 1999, 401:811-815. 21. Yang RB, Mark MR, Gray A, Huang A, Xie MH, Zhang M, Goddard A, Wood WI, Gurney AL, Godowski PJ: Toll-like receptor-2 mediates lipopolysaccharide-induced cellular signalling. Nature 1998, 395:284-288. 22. Kirschning CJ, Wesche H, Merrill Ayres T, Rothe M: Human toll-like receptor 2 confers responsiveness to bacterial lipopolysaccharide. J Exp Med 1998, 188:2091-2097. 23. Shimazu R, Akashi S, Ogata H, Nagai Y, Fukudome K, Miyake K, Kimoto M: MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4. J Exp Med 1999,189:1777-1782. 24. Chow JC, Young DW, Golenbock DT, Christ WJ, Gusovsky F: Tolllike receptor-4 mediates lipopolysaccharide-induced signal transduction. J Biol Chem 1999, 274:10689-10692. 25. Means TK, Wang S, Lien E, Yoshimura A, Golenbock DT, Fenton MJ: Human toll-like receptors mediate cellular activation by Mycobacterium tuberculosis. J Immunol 1999, 163:3920-3927. 26. Heine H, Kirschning CJ, Lien E, Monks BG, Rothe M, Golenbock DT: Cutting edge: cells that carry a null allele for toll-like receptor 2 are capable of responding to endotoxin. J Immunol 1999, 162:6971-6975. 27. Kawai T, Adachi O, Ogawa T, Takeda K, Akira S: Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity 1999, 11:115-122.