Extracellular Toll-Like Receptor 2 Region Containing Ser 40 -Ile 64 but Not Cys 30 -Ser 39 Is Critical for the Recognition of Staphylococcus aureus Peptidoglycan*

Toll-like receptor 2 (TLR2) and CD14 function as pattern recognition receptors for bacterial peptidoglycan (PGN). TLRs and CD14 possess repeats of the leucine-rich motif. To address the role of the extracellular domain of TLR2 in PGN signaling, we constructed CD14/ TLR2 chimeras, in which residues 1–356 or 1–323 of CD14 were substituted for the extracellular domain of TLR2, and five deletion mutants of TLR2, in which the progressively longer regions of extracellular TLR2 regions were deleted. PGN induced NF- (cid:1) B activation in HEK293 cells expressing TLR2 but not in cells expressing CD14/TLR2 chimeras. The cells transfected with a deletion mutant TLR2 (cid:2) Cys30-Ile64 as well as TLR2 (cid:2) Cys30-Asp160 and TLR2 (cid:2) Cys30-Asp305 failed to respond to PGN, indicating the importance of the TLR2 region Cys 30 -Ile 64 . Although TLR2 (cid:2) Cys30-Ser39 conferred cell responsiveness to PGN,

Gram-negative bacteria, can elicit the excessive release of proinflammatory cytokines from immune cells, resulting in the reproduction of most clinical manifestations of bacterial infections (2)(3)(4).
Membrane-bound CD14 (mCD14), a glycosylphosphatidylinositol (GPI)-anchored 55-kDa protein with the peptide of Thr 1 -Ala 356 on the surfaces of macrophages and polymorphonuclear leukocytes, serves as a cell-activating receptor for bacterial PGN as well as for LPS (5,6). CD14 avidly binds PGN as well as LPS (7). CD14 also exists in a soluble form (sCD14) in serum (8). Recombinant sCD14 consisting of the CD14 region Thr 1 -Gly 323 has been demonstrated to activate cells in response to LPS (9). Because mCD14 is a GPI-linked membrane receptor but lacks the intracellular signaling domain, signal transducing receptors for LPS and PGN have been postulated.
Toll, originally identified as a regulator for dorsoventral polarity of the Drosophila embryo (10), plays critical functions in the host defense mechanism of the adult fly (11). Toll-like receptors (TLRs), mammalian homologues of Toll, have recently been identified and implicated in signaling by LPS and PGN (12). TLRs, unlike CD14, possess a cytoplasmic domain homologous to that of interleukin-1 receptor (IL-1R). Stimulation by LPS and PGN through TLRs initiates an IL-1R-like NF-B signaling cascade (13). Genetic approaches have revealed that TLR4 is a signaling receptor for LPS. Pro 712 3 His mutation in Tlr4 gene causes defective response to LPS in C3H/HeJ mice (14). TLR4-knockout mice were hyporesponsive to LPS (15), confirming that TLR4 is a receptor responsible for LPS signaling. Macrophages from TLR2-deficient mice produced pro-inflammatory cytokines in response to LPS but were hyporesponsive to Staphylococcus aureus PGN (16,17). Recent in vitro studies also demonstrate that PGN-induced cell activation is mediated by TLR2 (18,19).
Accumulating evidence has revealed that CD14 and TLR2 function as pattern recognition receptors (5,20). The structural characteristic of these proteins is that they possess leucine-rich repeats (LRRs) (21), which appear to be involved in proteinprotein interaction (22,23). Individual repeats are predicted to contain a short ␤-strand and ␣-helix approximately antiparallel to each other (22,23). CD14 and TLR2 possess 10 and 19 LRRs, respectively. The amino-terminal half of the CD14 molecule, including the region between amino acids 51 and 64 is sufficient for the binding to PGN and LPS (24 -27). It remains unknown which region of the amino-terminal extracellular domain of TLR2 is involved in PGN recognition. To address the role of the extracellular domain of TLR2 in PGN signaling, we constructed CD14/TLR2 chimeras, in which mCD14 (residues 1-356) or sCD14 (residues 1-323) was substituted for the extracellular domain of TLR2, and five deletion mutants of TLR2, in which the progressively longer regions of extracellular TLR2 regions were deleted. This study demonstrates the importance of the amino-terminal region of TLR2 containing Ser 40 -Ile 64 in PGN signaling.
DNA Constructs-The 1.3-kb cDNA for human CD14 was obtained as described previously (28). The 2.6-kb cDNA for human TLR2 was obtained by reverse transcriptase-polymerase chain reaction (RT-PCR) using RNA isolated from U937 cells.
Transfection-HEK293 cells were transfected with expression plasmids by FuGENE 6 Transfection Reagent (Roche Molecular Biochemicals, Indianapolis, IN) according to the manufacturer's instructions.
Immunoblot-The transfected HEK293 cells were washed twice with ice-cold PBS and subsequently washed once with lysis buffer (10 mM Hepes, pH 7.4, containing 1.5 mM MgCl 2 , and 10 mM KCl). The cells were scraped with 1 ml of lysis buffer and were transferred into centrifuge tubes. After the centrifugation at 14,000 rpm for 15 min, the pellet was suspended with 40 l of lysis buffer containing 2.5% (v/v) Nonidet P-40. The suspension was then incubated on ice for 10 min and centrifuged at 14,000 rpm for 10 min. The supernatant containing the cytosolic and membrane proteins was finally collected. Protein concentration was estimated by the bicinchoninic assay (BCA, Pierce) using bovine serum albumin as a standard. 20 g of the supernatant proteins was subjected to SDS-polyacrylamide (10%) gel electrophoresis by the method of Laemmli (30). The proteins on the gel were electrotransferred onto a polyvinylidene difluoride (PVDF) membrane. Nonspecific binding was blocked by incubating the membrane with PBS containing 3% (w/v) skim milk and 0.1% (v/v) Triton X-100 (blocking buffer) for 30 min. The membrane was then incubated with 10 g/ml anti-CD14 polyclonal rabbit IgG (28) or 2 g/ml goat antibody against the TLR2 carboxyl terminus (Santa Cruz Biotechnology, Santa Cruz, CA) for 100 min. After washing the membranes with the blocking buffer three times, they were incubated for 1 h with HRP-labeled goat anti-rabbit IgG (1:2000, Bio-Rad, Hercules, CA) for CD14 and HRP-labeled donkey anti-goat IgG (1:500, Santa Cruz Biotechnology) for TLR2. The membranes were finally washed six times with PBS containing 0.1% (v/v) Triton X-100. The proteins that reacted with the antibodies were visualized by using a chemiluminescence reagent (SuperSignal, Pierce).
NF-B Reporter Assay-Activation of NF-B was measured as previously described (13,18). HEK293 cells were plated at 1 ϫ 10 5 cells per well in 24-well plates on the day before transfection. The cells were transiently transfected by FuGENE 6 transfection reagent (Roche Molecular Biochemicals, Indianapolis, IN) with 30 ng of an NF-B reporter construct (pNF-kB-Luc, Stratagene) and 3.5 ng of a construct-directing expression of Renilla luciferase under the control of the constitutively active thymidine kinase promoter (pRL-TK, Promega), together with 150 ng of each transfectant gene of CD14, TLR2, chimeras, and deletion mutants. Forty-eight hours after transfection, the cells were stimulated with PGN for 6 h in the absence of serum unless otherwise indicated and luciferase activity was measured by using the Dual-Luciferase reporter assay system (Promega) according to the manufacturer's instructions.
Effect of Synthetic Peptide on PGN-induced NF-B Activation-A peptide (peptide 40 -64), SGSLNSIPSGLTEAVKSLDLSNNRI, corresponding to the amino acid residues Ser 40 -Ile 64 of TLR2 and a control peptide, GTPVNYTNWYRG, were synthesized by Sigma Genosys Japan and purified by high-performance liquid chromatography. PGN was preincubated at 37°C for 30 min in the presence of 0.1-10 g/ml peptide 40 -64 or control peptide in serum-free medium. After the preincubation, a mixture of the peptide and PGN was added to the wells in which HEK293 cells had been transfected with TLR2 cDNA in pcDNA3.1(ϩ) plasmid vector. After incubation at 37°C for 6 h under 5% CO 2 , NF-B activity was measured by luciferase assay as described above.
Stably Transfected Cell Lines-Stably transfected cell lines were established by retroviral vector system (31). The pCX4bsr plasmid vector used, which was a kind gift from Dr. T. Akagi (Osaka BioScience Institute, Japan), is a modified version of pCXbsr (31) lacking the internal initiation codons within the gag region.
Immunostaining of the CD14/TLR Chimeras and Fluorescence Microscopy-Stably transfected HEK293 cell lines expressing CD14, mCD14/TLR2 chimera, or sCD14/TLR2 chimera had been grown on polylysine-coated coverglasses. The cells were rinsed with PBS and incubated with 5 g/ml rhodamine-conjugated concanavalin A (Molecular Probes) in serum-free Dulbecco's modified Eagle's medium at 4°C for 15 min. After the incubation, the medium was removed and the cells were fixed in methanol for 4 min at Ϫ20°C. The cells were washed with 50 mM PIPES buffer (pH 7.2) containing 100 mM NaCl, 1 mM EGTA, 2 mM MgCl 2 , and 10% (w/v) Block Ace (Dainippon Pharmaceutical, Tokyo) (buffer A) and then incubated with 5 g/ml anti-CD14 monoclonal antibody (MY4, Coulter) diluted with buffer A at room temperature for 30 min. The cells were then washed with buffer A and further incubated with Alexa 488-conjugated anti-mouse IgG (1:200) at room temperature for 20 min. The cells were finally washed three times with buffer A, sealed in the presence of Vectashield Antifade (Vector Laboratories) and examined using a laser microscope (LSM510, Carl Zeiss, Tokyo, Japan) with a ϫ63 oil planapochromatic lens (NA1.4). Digital Images were acquired and processed using Adobe Photoshop, version 5.0 (Mountain View, CA) and CorelDRAW software (Corel Corp.).
Recombinant CD14 and PGN Binding-Recombinant CD14 (CD14 T1-G323 ), consisting of the CD14 region Thr 1 -Gly 323 and a sixhistidine tag, was expressed in Chinese hamster ovary cells, and the recombinant protein was purified from the culture medium by an affinity column of nickel-nitrilotriacetic acid beads as described previously (28).
A PGN binding study was carried out by the method described previously for the binding of CD14 to LPS (28). 5 and 20 g of PGN in 20 l of ethanol was added onto microtiter wells (Immulon 1B, Dynex), and the solvent was evaporated in ambient air. After nonspecific binding to the wells was blocked with 10 mM Hepes buffer (pH 7.4) containing 0.15 M NaCl, 5 mM CaCl 2 , and 5% (w/v) bovine serum albumin (buffer B), 0 -5 g/ml CD14 T1-G323 (50 l/well) in the buffer B was added and incubated at 37°C for 6 h. The wells were then washed with PBS containing 3% (w/v) skim milk and 0.1% (v/v) Triton X-100 and incubated with anti-CD14 IgG, followed by incubation with horseradish peroxidase-conjugated goat anti-rabbit IgG. The peroxidase reaction was finally performed using o-phenylenediamine as a substrate after washing the wells with PBS containing 0.1% (v/v) Triton X-100. The binding of CD14 T1-G323 to PGN was detected by measuring absorbance at 492 nm.
To determine whether CD14, TLR2, and the CD14/TLR2 chimeras were expressed in HEK293 cells after transient transfection, mRNA and protein expressions were analyzed by using RT-PCR and immunoblotting. RT-PCR analysis revealed that the PCR products exhibited the expected mobilities of bands at 1221 bp for CD14, 2600 bp for TLR2, 1799 bp for sCD14/TLR2 chimera, and 1898 bp for mCD14/TLR2 chimera ( Fig. 2A). The results indicate that mRNAs for these four proteins were expressed in HEK293 cells after transient transfection. The protein expression was also confirmed by immunoblotting analysis with the anti-carboxyl terminus of TLR2 antibody (Fig. 2B) and with anti-CD14 antibody (Fig. 2C). CD14 and TLR2 exhibited mobilities at ϳ55 and 90 kDa, respectively, as described previously (5,32). The CD14/TLR2 chimeras reacted with both anti-TLR2 and anti-CD14 antibodies and exhibited bands at ϳ70 -75 kDa. These results clearly show that the proteins of CD14, TLR2, and CD14/TLR2 chimeras were expressed in transiently transfected-HEK293 cells.
CD14 Cannot Functionally Replace the Extracellular Domain of TLR2 in PGN Signaling-Both CD14 and TLR2, which possess leucine-rich repeats in the extracellular domains, have been proposed to be pattern recognition receptors for PGN (5,20). CD14 binds PGN (7), and TLR2 confers cell responsiveness to PGN (18). The soluble form of CD14, consisting of the region Met 1 -Gly 323 , is biologically active and is capable of binding LPS (9). To investigate the roles of the extracellular domains of CD14 and TLR2 in PGN signaling, we constructed CD14/TLR2 chimeras containing the CD14 region of Met 1 -Ala 356 or Met 1 -Gly 323 and estimated NF-B activation using luciferase reporter gene assay in transiently transfected HEK293 cells. PGN stimulated NF-B reporter activity in wt TLR2-but not CD14-transfected cells in a manner dependent upon PGN concentrations (Fig. 3). In addition, cotransfection of CD14 cDNA with TLR2 cDNA enhanced cellular responses by ϳ43-200%. However, the cells transfected with sCD14/TLR2 chimera cDNA or mCD14/TLR2 chimera cDNA failed to respond to any  2. Expression of CD14, TLR2, and CD14/TLR2 chimeras in transfected HEK293 cells. A, 48 h after transfection with CD14, TLR2, mCD14/TLR2 chimera, or sCD14/TLR2 chimera, RNA was isolated from HEK293 cells and RT-PCR for each protein and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was performed as described under "Experimental Procedures." B and C, the cells transfected were lysed with the buffer containing Nonidet P-40, and the cytosolic and membrane proteins were isolated as described under "Experimental Procedures." 20 g of the proteins was analyzed by 10% polyacrylamide gel and transferred onto the PVDF membranes. The membranes were probed with anti-TLR2 antibody (B) or anti-CD14 antibody (C), and the proteins reacting with the antibodies were visualized by using a chemiluminescence reagent as described under "Experimental Procedures." doses of PGN (Fig. 3). We performed the control experiments in which the reporter plasmids were omitted or the reporter plasmids were transfected without the tested plasmids. In these cases none of the cells exhibited significant luciferase activities when stimulated with PGN, indicating that the responses in TLR2 and/or CD14-transfected cells are dependent on the presence of the NF-B reporter plasmids. When the chimera-transfected cells were examined in the presence of serum, no cellular activation to PGN was observed. Cotransfection of mCD14 with the chimeras also failed to confer NF-B activation to PGN. We also examined the effect of chimeras on PGN signaling in the cells expressing wt TLR2. When sCD14/TLR2 or mCD14/TLR2 chimera cDNA was cotransfected with wt TLR2 cDNA, the activities of measured NF-B at 100 ng/ml PGN were enhanced to 34 or 274% greater (mean of three or four experiments), respectively, than those activities in cells transfected with wt TLR2 cDNA alone. These results confirmed that CD14 cannot directly transmit PGN signaling but enhances cellular responses through TLR2 and showed that these chimeras can enhance PGN signaling mediated by TLR2, although they cannot directly transmit PGN signaling, indicating that they possess the same biological activities as mCD14. The present data demonstrate that CD14 cannot functionally replace the extracellular domain of TLR2 in PGN signaling.
Cell Surface Localization of CD14/TLR2 Chimeras-Because cell surface localization of CD14/TLR2 chimeras is of critical importance for the interpretation of the results shown in Fig. 3, we attempted to define the cellular localization of the chimeras in stably transfected HEK293 cell lines. For this purpose the HEK293 cell line expressing CD14, mCD14/TLR2 chimera, or sCD14/TLR chimera was labeled with rhodamineconjugated concanavalin A and immunostained with anti-CD14 monoclonal antibody, and it was examined by a laser scanning confocal microscope (Fig. 4). Each protein expressed localized, mainly at the cell surface membrane, and colocalized with concanavalin A used as an established marker for cell surface glycoproteins. These results demonstrate the cell surface localization of the chimeric proteins.

PGN-induced NF-B Activation in Stably Transfected
Cell Lines Expressing the Chimeras-To confirm the results obtained from the transiently transfected cells, the stably transfected cell lines expressing the chimeras were also examined for NF-B activation in response to PGN (Fig. 5). PGN stimu-lated NF-B reporter activity in the cells expressing wt TLR2 but not in the cells expressing CD14 in a manner dependent upon PGN concentration. The cells expressing mCD14/TLR2 or sCD14/TLR2 chimera neither responded to PGN. Taken together, we conclude that the CD14/TLR2 chimeras cannot transmit PGN signaling. The data shown in Figs. 4 and 5 confirm that CD14 cannot functionally replace the extracellular domain of TLR2 in PGN signaling.
Binding of CD14 T1-G323 to PGN-To determine whether the extracellular domains of CD14/TLR2 chimeras bind to PGN, we constructed recombinant CD14 (CD14 T1-G323 ) consisting of the CD14 region Thr 1 -Gly 323 , because the CD14/TLR2 chimeras contain this CD14 region as the extracellular domain. The recombinant CD14 T1-G323 migrated as bands of 46 -56 kDa when analyzed by electrophoresis under denaturing and reducing conditions (Fig. 6A), as described previously (28). We next performed the direct binding study using PGN coated onto microtiter wells (Fig. 6B). The CD14 T1-G323 bound to PGN in manners dependent upon the CD14 concentrations and the amounts of PGN coated on the well. These results indicate that the CD14 region of Thr 1 -Gly 323 directly binds to PGN and confirm the previously demonstrated binding of CD14 to PGN (7). The data also suggest that the extracellular CD14 region of the chimeric molecules can bind PGN.
Characterization and Expression of TLR2 Deletion Mutants-We next constructed five deletion mutants of TLR2 (Fig.  1B), in which the progressively longer regions of the extracellular domain were deleted, to investigate the essential region of TLR2 extracellular domain in PGN recognition. mRNA and protein expressions were analyzed by using RT-PCR and immunoblotting. Each deletion mutant exhibited the expected molecular sizes of PCR product at 1746 -2573 bp (Fig.  7A). Next, we examined whether each mutant protein was expressed in HEK293 cells transiently transfected with each mutant cDNA. The mutant proteins were recognized by the antibody against the TLR2 carboxyl terminus and showed apparent molecular sizes of 50 -90 kDa when analyzed by immunoblotting (Fig. 7B). These results clearly indicate that mRNAs and the proteins of the deletion mutants were expressed in transfected HEK293 cells.
NF-B Activation in HEK293 Cells Transiently Transfected with TLR2 Deletion Mutants-To investigate which region of the amino-terminal extracellular domain of TLR2 is required for PGN signaling, we examined cellular responses to PGN in the cells transfected with the deletion mutant cDNAs by measuring NF-B activation. Three deletion mutants (TLR2 ⌬Cys30-Ile64 , TLR2 ⌬Cys30-Asp160 , and TLR2 ⌬Cys30-Asp305 ) failed to induce NF-B activation when stimulated with 100 -10000 ng/ml PGN (Fig. 8A), indicating that the TLR2 region of Cys 30 -Ile 64 is essential for PGN signaling. To further examine the critical stretch in the TLR2 region of Cys 30 -Ile 64 , two deletion mutants (TLR2 ⌬Cys30-Ser39 and TLR2 ⌬Ser40-Ile64 ) were constructed and examined for cellular responses to PGN. The cells transfected with TLR2 ⌬Cys30-Ser39 responded to PGN. Although the responses to PGN in the cells expressing TLR2 ⌬Cys30-Ser39 appeared weaker than those in the cells expressing wt TLR2, the cells expressing TLR2 ⌬Cys30-Ser39 clearly retained a significant ability to induce NF-B activation (Fig. 8A). The measured activity of NF-B reporter assay induced by TLR2 ⌬Cys30-Ser39 was 23% lower than that observed for wt TLR2 at 10,000 ng/ml PGN. However, the cells transfected with TLR2 ⌬Ser40-Ile64 exhibited no responses to any concentrations of PGN. Because cotransfection of mCD14 with TLR2 enhances cellular responses to PGN (18), we tested the abilities of the deletion mutant-transfected cells to induce NF-B activation in the presence of mCD14 (Fig. 8B). The results obtained were essentially the same as those obtained without cotransfection with mCD14. The cells transfected with TLR2 ⌬Cys30-Ser39 as well as wt TLR2 responded well. However, cotransfection of mCD14 with TLR2 ⌬Cys30-Ile64 or TLR2 ⌬Ser40-Ile64 failed to confer cellular responsiveness to PGN (Fig. 8B). These results demonstrate that the TLR2 region of Cys 30 -Ser 39 is not required for PGN recognition and that the TLR2 region containing Ser 40 -Ile 64 is critical for bestowal of cell responsiveness to PGN.
Effect of Synthetic Peptide on NF-B Activation Induced by PGN-To confirm the importance of the TLR2 region of Ser 40 -Ile 64 , we prepared a synthetic peptide (peptide 40 -64) corresponding to amino acid residues Ser 40 -Ile 64 of TLR2. The peptide was tested to see whether it competed with TLR2  7. Expression of wt TLR2 and TLR2 deletion mutants in transfected HEK293 cells. A, 48 h after transfection with wt TLR2 or each deletion mutant, RNA was isolated from HEK293 cells and RT-PCR for each protein and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was performed as described under "Experimental Procedures." B, the transfected cells were lysed with the buffer containing Nonidet P-40, and the cytosolic and membrane proteins were isolated as described under "Experimental Procedures." 20 g of the proteins was analyzed by 10% polyacrylamide gel and transferred onto the PVDF membranes. The membranes were probed with anti-TLR2 antibody, and the proteins that reacted with the antibodies were visualized by using a chemiluminescence reagent as described under "Experimental Procedures." expressed in HEK293 cells for PGN recognition. Control peptide at 10 g/ml did not affect NF-B activation when stimulated by PGN. When the peptide 40 -64 was incubated with PGN and the TLR2-transfected cells, the NF-B activities were significantly reduced (Fig. 9). NF-B activities observed in the presence of 5 and 10 g/ml peptide 40 -64 were decreased to 50 -63% of those observed for control peptide. These results further support the idea that the TLR2 region containing Ser 40 -Ile 64 is critical for PGN recognition. DISCUSSION The characteristics of mice harboring null allele for TLR2 (16,17) provide strong in vivo evidence that TLR2 is not involved in the LPS response but is indeed involved in cellular responses to PGN. In vitro study with overexpression experiments confirms the essential role of TLR2 in PGN signaling (18). C3H/HeJ mice that have been shown to be hyporesponsive to LPS (14) have been found to possess a missense point mutation within the intracellular domain of TLR4 resulting in the substitution of His 712 for Pro. This proline residue is highly conserved among the TLR family, including TLR2. The cytoplasmic domains of TLRs homologous to IL-1R have been demonstrated to be essential for signaling leading to NF-B activation (13). Because each TLR is involved in signaling for different bacterial ligands (13, 16, 18 -20, 33, 34), the extracellular domains of TLRs may define the specificities for various ligands. This study focused on the extracellular domain of TLR2 and investigated the structural requirements of the amino-terminal TLR2 for PGN recognition. The results clearly demonstrate that CD14 cannot functionally replace the extracellular domain of TLR2 in PGN signaling, that the TLR2 region of Cys 30 -Ser 39 is not required for PGN recognition, and that the TLR2 region containing Ser 40 -Ile 64 is critical for PGN signaling.
CD14 serves as a functional receptor for LPS and PGN (5,7,35). TLRs have also been demonstrated to be responsible for cellular responsiveness to LPS and PGN (13,18). Both receptors possess homologous structures consisting of leucine-rich repeats (LRRs) characteristic of a short ␤-strand and ␣-helix (23). Although CD14 avidly binds LPS and PGN, CD14 alone cannot confer cell responsiveness to bacterial ligands, because CD14 does not possess the intracellular signaling domain. In addition to membrane-bound CD14 (mCD14), consisting of Thr 1 -Ala 356 , the soluble form of CD14 (sCD14) is also able to activate cells in response to bacterial ligands (35). Recombinant sCD14, consisting of the CD14 region Thr 1 -Gly 323 , has been proved to be active (9). We constructed two chimeric molecules, in which mCD14 (Thr 1 -Ala 356 ) or sCD14 (Thr 1 -Gly 323 ) was substituted for the extracellular TLR2 domain (mCD14/TLR2 chimera or sCD14/TLR2 chimera, respectively). Both chimeras were unable to activate the cells in response to PGN regardless of the presence of coexpressed mCD14. Cotransfection of the chimera cDNA with TLR2 cDNA enhanced NF-B activation when compared with the transfection with TLR2 cDNA alone, indicating that these chimeras possess the same biological activities as mCD14. In addition, the CD14/TLR2 chimeras localized at the cell surface membrane (Fig. 4). These data may rule out the possibility that these chimeras exhibit global defects in This study has shown that the CD14/TLR2 chimeras cannot activate the cells in response to PGN using the transiently transfected cells and the stably transfected cell lines (Figs. 3 and 5). The results demonstrate that CD14 cannot functionally replace the extracellular TLR2 domain in PGN signaling. We further attempted to differentiate between PGN binding and signaling after PGN binding. To address this issue, we determined whether the CD14 region Thr 1 -Gly 323 (CD14 T1-G323 ) of the extracellular domains of CD14/TLR2 chimeras bound PGN in a cell free assay system, although we were unable to use the whole molecules of the chimeras. CD14 T1-G323 avidly bound to PGN (Fig. 6), suggesting that the extracellular domains of the chimeras can bind PGN. In addition, cell surface localization of the chimeric proteins has been demonstrated by immunostaining with anti-CD14 antibody (Fig. 4). Thus, it is possible to presume that the CD14/TLR2 chimeras can bind PGN but cannot transmit its signals.
We then focused on the TLR2 region of Cys 30 -Ile 64 , because the deletion mutant TLR2 ⌬Cys30-Ile64 as well as the mutants with longer deletion failed to induce NF-B activation in response to PGN. The further experiments with TLR2 ⌬Cys30-Ser39 and TLR2 ⌬Ser40-Ile64 narrowed the critical stretch of primary sequence in the amino-terminal TLR2 region. Because TLR2 ⌬Cys30-Ser39 conferred significant cell responsiveness, albeit to a lesser extent compared with wt TLR2, it is concluded that the TLR2 region of Cys 30 -Ser 39 is not required for PGN recognition. The loss of the cell responsiveness in the cells expressing TLR2 ⌬Ser40-Ile64 demonstrates the importance of the TLR2 region Ser 40 -Ile 64 . Because a criticism might be raised that the conformational changes due to the 25-amino acid deletion caused the receptor to be inactive, we examined whether the synthetic peptide corresponding to the TLR2 region of Ser 40 -Ile 64 competed with wt TLR2 for PGN signaling. The peptide was not able to block completely the NF-B activation, but nonetheless significantly attenuated it. Thus, it is possible to conclude that the TLR2 region contiguous to Ser 40 -Ile 64 plays a critical part in PGN recognition, although the activity of the TLR2 region Ser 40 -Ile 64 does not explain all the activity of TLR2 in the recognition of PGN. The TLR2 region of Ser 40 -Ile 64 contains the portion of the first sequence of the leucinerich motif ( 54 VXXLXLXXNXIXXIXXXXLXX 74 ), which appears to be a consensus sequence seen in a "typical" subfamily according to the classification of the leucine-rich repeat (LRR) protein superfamily described by Kajava (22). Repeats of the leucine-rich motif and a stretch of the sequence preceded by this motif may be important for the specific recognition of bacterial components.
TLR2 contains 19 repeats of the leucine-rich motif. The proteins containing LRRs appear to be involved in protein-protein interaction (22,23). The structure of this unique motif adopts a curved shape resembling a horseshoe with an ␣-helix lining its outer circumference and ␤-strands forming a parallel ␤-sheet along its inner circumference (22). The LRR proteins, including CD14, often possess flanking cysteine clusters at the amino terminus (36). TLR2 exhibits Cx5C in the region of Cys 30 -Ser 39 . Although the role of this flanking cysteine cluster at the TLR2 amino terminus remains unknown, the data obtained from TLR2 ⌬Cys30-Ser39 clearly demonstrate that the cysteine-rich cluster (Cys 30 -Cys 36 ) of TLR2 is not required for PGN recognition.
In conclusion, this study identified a stretch of the sequence of the amino-terminal TLR2 extracellular domain that is required for PGN recognition. The TLR2 region of Cys 30 -Ser 39 is not required, but the region containing Ser 40 -Ile 64 is critical for PGN signaling.