Cross Talk between MyD88 and Focal Adhesion Kinase Pathways1
http://www.100md.com
免疫学杂志 2005年第11期
Abstract
Focal adhesion kinase (FAK) is a nonreceptor protein tyrosine kinase involved in signaling downstream of integrins, linking bacterial detection, cell entry, and initiation of proinflammatory response through MAPKs and NF-B activation. In this study, using protein I/II from Streptococcus mutans as a model activator of FAK, we investigated the potential link between FAK and TLR pathways. Using macrophages from TLR- or MyD88-deficient mice, we report that MyD88 plays a major role in FAK-dependent protein I/II-induced cytokine release. However, response to protein I/II stimulation was independent of TLR4, TLR2, and TLR6. The data suggest that there is a cross talk between FAK and MyD88 signaling pathways. Moreover, MyD88-dependent, LPS-induced IL-6 secretion by human and murine fibroblasts required the presence of FAK, confirming that MyD88 and FAK pathways are interlinked.
Introduction
Bacterial recognition implicates a wide variety of pattern recognition receptors specific for pathogen-associated microbial patterns (PAMPs), 5 including TLRs, lectins, scavenger receptors, and integrins. Among them, TLRs are critical for the development of innate immune responses (1). TLR signaling is mediated by cytoplasmic Toll/IL-1R (TIR) domain homo- or heteromeric associations with TIR-containing adaptors (2). MyD88, an adaptor protein used by all TLRs except TLR3, activates MAPKs and NF-B, leading to proinflammatory cytokine expression. MyD88-independent pathways, involving the TIR domain-containing adaptor inducing IFN- (TRIF) are used by TLR4 and TLR3, together with TRIF-related adaptor molecule in the case of TLR4. TRIF pathway activates IFN regulatory factor 3, leading to the synthesis of IFN- and -. Moreover, receptor interacting protein 1, a component of the TNFR signaling pathway, associates with the C-terminal region of TRIF and mediates TLR4 and TLR3-NF-B late activation (3). Thus, various PAMP/pattern recognition receptor interactions result in MAPK and NF-B activation and consequent expression of proinflammatory cytokines, and there is now strong evidence that many intracellular signaling molecules are shared among the activated signaling pathways. However, the kinases involved in these pathways are not yet fully elucidated.
Focal adhesion kinase (FAK), a nonreceptor protein tyrosine kinase involved in signaling downstream of integrins (4), was shown recently to trigger inflammatory responses. Protein I/II, a cell wall component from oral streptococci, binding to integrin 51 induces the production of inflammatory mediators such as IL-6 and IL-8 by human monocytes, epithelial cells, endothelial cells, and fibroblasts (5, 6). The signaling events leading to this cytokine release involve FAK, ERK1/2, and JNKs as well as AP-1-binding activity and nuclear translocation of NF-B (6, 7). FAK is an interesting candidate to link bacterial detection, initiation of proinflammatory responses, and cell entry, because FAK is involved in invasin-mediated bacterial uptake (8). Based on these observations, we investigated the possibility of a cross talk between the integrin/FAK and TLR pathways.
Using protein I/II from Streptococcus mutans as a model activator of FAK, together with macrophages from TLR- or MyD88-deficient mice, we show that MyD88 and FAK, but neither TLR2, TLR4, nor TLR6, are involved in the response to protein I/II that leads to cytokine release. This cross talk between FAK and MyD88 seems to be a general phenomenon in proinflammatory cytokine response, because LPS-induced cytokine release also depends on the presence of both FAK and MyD88, as evidenced in primary human and murine cells.
Materials and Methods
Cytokine ELISA
Cell supernatants were assayed for cytokine content using commercially available ELISA reagents for TNF-, IL-6, and IL-8 (Duoset; R&D Systems).
Nitrite measurements
Nitrite concentrations in cell supernatants were determined using the Griess reaction (3% phosphoric acid, 1% p-aminobenzenesulfonamide, 1% N-(1-napthyl)ethylenediamide) as previously described (14).
Results
Protein I/II, a ligand of integrin 51, induces FAK phosphorylation and TNF- and NO production
We showed previously that protein I/II, a ligand of integrin 51, induces cytokine production by human endothelial and epithelial cells and FLSs and that this effect is FAK dependent (5, 6). In this study, we first verified the capacity of protein I/II to stimulate phosphorylation of FAK in murine bone marrow-derived macrophages. Cell lysates were analyzed directly by blotting with specific anti-FAK (pY397) Abs. Stimulation with protein I/II resulted in an increasing amount of phosphorylated FAK (Fig. 1A), which was detectable within 5 min and remained elevated for at least 30 min. In addition, protein I/II induced a strong release of TNF- and NO by activated macrophages, reaching levels similar to those obtained after endotoxin stimulation (Fig. 1, B and C, wild-type).
The TLR2 and TLR4 independence of protein I/II activation was then confirmed in human FLSs, a cell population known to contribute to joint inflammation in rheumatoid arthritis patients. Pretreatment with Abs directed against TLR4 or TLR2 had essentially no effect on protein I/II-induced IL-8 release by FLSs (data not shown).
Taken together, these data demonstrate that TLR2, TLR4, and TLR6 are not involved in protein I/II-induced TNF- and NO release by activated murine macrophages and in protein I/II-induced IL-6 and IL-8 release by activated human FLSs.
Protein I/II-induced FAK phosphorylation is independent of MyD88
We then tested the role of MyD88 in FAK phosphorylation. Stimulation of MyD88–/– macrophages with protein I/II resulted in an increasing amount of phosphorylated FAK (Fig. 3), which was detectable within 5 min and maximal after 15 min of stimulation, comparable to what was shown in wild-type cells (Fig. 1A). Therefore, although MyD88 is involved in the response to protein I/II that leads to cytokine release, FAK phosphorylation seems to be largely independent of the presence of MyD88, suggesting that phosphorylation of FAK occurs upstream of MyD88 or that both pathways are parallel.
LPS stimulation induces FAK phosphorylation in macrophages and FLSs
Our data suggesting that MyD88 and FAK are both necessary in the signaling pathway induced by protein I/II interaction with integrin 51, we next investigated whether this cross talk between FAK and MyD88 was a general phenomenon by assessing the role of FAK in signaling pathways known to involve MyD88. Cytokine response to LPS is largely MyD88 dependent, and we examined the capacity of LPS to induce FAK phosphorylation in murine macrophages. After stimulation, cell lysates were analyzed directly by blotting with specific anti-FAK (pY397) Abs as above. Stimulation by LPS resulted in an increasing amount of phosphorylated FAK (Fig. 4A), which was detectable within 5 min and remained elevated for at least 30 min. Similar FAK phosphorylation was obtained after LPS stimulation of human FLSs (Fig. 4B). Thus, LPS induced FAK phosphorylation in murine and human primary cells.
LPS-induced IL-6 release occurs via a FAK-dependent pathway
We next asked whether FAK could contribute to the signaling events leading to cytokine release from LPS-activated cells. First, using FAK+/+ and FAK–/– primary mouse embryo fibroblasts, we showed that LPS induced IL-6 release by FAK+/+ fibroblasts, but there was essentially no IL-6 production by FAK–/– fibroblasts (p < 0.01). This suggested that FAK is involved in LPS-induced release of IL-6. To rule out the possibility that the lack of IL-6 release could be due to a nonspecific inhibition of other cellular functions, FAK+/+ fibroblasts were transfected with FAK-related nonkinase (FRNK), the C-terminal region of FAK, which is known to block FAK activation when overexpressed. FAK+/+ fibroblasts transfected with a GFP-expressing vector were used as a control. As shown in Fig. 5A, overexpression of FRNK strongly inhibited IL-6 release from LPS-activated FAK+/+ cells (p < 0.01), confirming the involvement of FAK in LPS-induced IL-6 secretion.
Taken together, our results show that MyD88 is involved in the response to protein I/II-integrin 51 interaction that leads to cytokine release through FAK phosphorylation, and this cross talk between the MyD88 and FAK pathways seems to be a general phenomenon in proinflammatory cytokine response, because MyD88-dependent, LPS-induced IL-6 secretion also depends on the presence of FAK.
Discussion
In the present study, we first demonstrated that absence of MyD88 abrogates or severely impairs TNF- and NO production by protein I/II-activated macrophages. These data suggest that MyD88, a key downstream adaptor of the TLR pathway, plays a major role in protein I/II-integrin-induced cytokine release. We showed previously, using FAK–/– and FRNK-transfected cells, that FAK is essential in IL-6 and IL-8 release by protein I/II-activated FLSs (6). Thus, we examined next how FAK and MyD88 pathways may interact. Two hypotheses were formulated, namely, protein I/II may activate MyD88 by interacting with a member of the TLR family, or alternatively, protein I/II might trigger a signaling pathway involving an intracellular cross talk between FAK and MyD88. To address these questions, we first used macrophages isolated from TLR2–/– and TLR6–/– mice insofar as these TLRs recognize various components from Gram-positive bacteria. TLR4, the receptor of LPS from Gram-negative bacteria that also interacts with fibronectin, a ligand of integrin 51, was also tested. Our results clearly show that TLR2, TLR4, and TLR6 are not essential for the stimulation of macrophages by protein I/II. Also, Abs to TLR4 and TLR2 failed to inhibit protein I/II-induced IL-8 release by human FLSs (data not shown). These data are in agreement with the observation that macrophage activation by group B streptococci requires MyD88 but is independent of TLR1, -2, -4, and -6 (17). Thus, we excluded TLR2, TLR4, and TLR6 as potential receptors for protein I/II.
We next showed that FAK plays an important role downstream of TLRs, because cytokine response to LPS is totally abrogated in FAK–/– cells. Furthermore, FAK-expressing cells transfected with the FAK inhibitor FRNK produced significantly lower amounts of IL-6 upon LPS challenge. This is in line with previous results showing that LPS induced FAK phosphorylation in a murine monocytic cell line (18). We further demonstrated that other TLR ligands may also induce FAK phosphorylation at Tyr397 and require FAK for cytokine release because comparable results have been obtained with the TLR2 ligand Pam3CSK4 (data not shown). Also, our findings extend to FAK the observation that TLR2 and TLR4 agonists induced tyrosine phosphorylation of the proline-rich tyrosine kinase 2 (Pyk2), which in turn increases tyrosine phosphorylation of paxillin, an adaptor protein involved in integrin signaling and binding to Pyk2, vinculin, and FAK, through MyD88-dependent and -independent pathways (19). However, in our experiments, MyD88 does not seem to be essential for FAK autophosphorylation because this kinase is phosphorylated in MyD88–/– macrophages activated with either LPS or protein I/II. Thus, FAK phosphorylation seems to occur upstream of MyD88, or in parallel to MyD88 recruitment, and FAK might be a key proximal effector of MyD88.
Recent studies have shown that PAMP stimulation leads to TLR recruitment to cholesterol-rich regions of the cell membrane such as lipid rafts (16). These specialized domains may contribute to the assembly of a receptor complex involving transmembrane proteins such as TLRs and integrins as well as intracellular proteins. Preliminary experiments suggest that LPS and protein I/II-induced cell activation is dependent on such lipid rafts. Indeed, disruption of raft-like domains with the cholesterol-binding drugs filipin and nystatin inhibits LPS and protein I/II-induced cytokine release, suggesting that receptor complex assembly in microdomains is essential for downstream signaling triggered by these two PAMPs (data not shown). Therefore, MyD88 and FAK may contribute to the intracytoplasmic signaling of this receptor complex.
MyD88 may be activated in at least two different ways. Ligand engagement of members of the IL-1R/TLR family induces the recruitment and phosphorylation of MyD88, but it has also been demonstrated that Fas ligation on macrophages enhances TLR4 signaling by suppressing the inhibitory interaction of Fas-associated death domain protein with MyD88 (20). One might speculate that FAK contributes to MyD88 phosphorylation, but FAK possesses neither a death domain nor a TIR domain for direct interaction with MyD88. Alternatively, FAK may participate in MyD88 activation by recruiting another kinase. In previous experiments, we addressed the involvement of Src kinases in protein I/II-induced signaling. Protein I/II did not induce phosphorylation of Src at Tyr418 (activated Src) nor mediate phosphorylation of FAK at Tyr925, a Src phosphorylation site (L. A. Neff, M. B. Zeisel, V. A. Druet, J. P. Klein, J. Sibilia, and D. Wachsmann, unpublished results). Thus, Src does not seem to play a role in protein I/II-induced signaling. It has long been known that protein tyrosine kinases are required for the induction of proinflammatory cytokines in response to LPS in macrophages. Jefferies et al. (21, 22) demonstrated Btk as a critical tyrosine kinase in LPS signaling in myeloid cells. Btk interacts with the TIR domain of TLRs as well as with downstream molecules such as MyD88, MAL, and IL-1R-associated kinase-1, but the precise nature of these interactions is not well defined. Btk is probably activated through Src kinases although other tyrosine kinases such as PI3K and Jak1 have also been shown to phosphorylate members of the Btk family (23). Because FAK has been demonstrated to be an upstream activator of Bmx/Etk in endothelial cells (24), it is conceivable that FAK might lead to the recruitment of MyD88 via the activation of a member of the Btk family of kinases and further downstream signaling to MAPKs, AP-1, and NF-B. Further studies of the involvement of additional kinase players in FAK and MyD88 pathway cross talk will be needed to complete our understanding of the molecular mechanisms linking bacterial detection, cell invasion, and proinflammatory response.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by grants from the French Ministry of Research, Pfizer, Région Alsace, Amgen, Caisse d’Assurance Maladie des Professions Libérales de Province, Wyeth Lederle, and Centre National de la Recherche Scientifique.
2 M.B.Z. and V.A.D. contributed equally to this work.
3 V.Q. and D.W. contributed equally to this work.
4 Address correspondence and reprint requests to Dr. Dominique Wachsmann, Faculté de Pharmacie, Université Louis Pasteur, 74 route du Rhin, 67400 Illkirch, France. E-mail address: wachs{at}pharma.u-strasbg.fr
5 Abbreviations used in this paper: PAMP, pathogen-associated microbial pattern; TIR, Toll/IL-1R; TRIF, TIR domain-containing adaptor inducing IFN-; FAK, focal adhesion kinase; FLS, fibroblast-like synoviocyte; Pam3CSK4, S-[2,3-bis(palmitoyloxy)-(2-RS)propyl]-N-palmitoyl-(R)-Cys-(S)-Ser-Lys4-OH, trihydrochloride; BLP, bacterial lipoprotein; FRNK, FAK-related nonkinase.
Received for publication January 5, 2005. Accepted for publication March 25, 2005.
References
Medzhitov, R.. 2001. Toll-like receptors and innate immunity. Nat. Rev. Immunol. 1: 135-145.
Takeda, K., S. Akira. 2004. Microbial recognition by Toll-like receptors. J. Dermatol. Sci. 34: 73-82.
Meylan, E., K. Burns, K. Hofmann, V. Blancheteau, F. Martinon, M. Kelliher, J. Tschopp. 2004. RIP1 is an essential mediator of Toll-like receptor 3-induced NF-B activation. Nat. Immunol. 5: 503-507.
Schlaepfer, D. D., C. R. Hauck, D. J. Sieg. 1999. Signaling through focal adhesion kinase. Prog. Biophys. Mol. Biol. 71: 435-478.[
Al-Okla, S., C. Chatenay-Rivauday, J. P. Klein, D. Wachsmann. 1999. Involvement of 51 integrins in interleukin 8 production induced by oral viridans streptococcal protein I/IIf in cultured endothelial cells. Cell. Microbiol. 1: 157-168.
Neff, L., M. Zeisel, V. Druet, K. Takeda, J. P. Klein, J. Sibilia, D. Wachsmann. 2003. ERK 1/2- and JNKs-dependent synthesis of interleukins 6 and 8 by fibroblast-like synoviocytes stimulated with protein I/II, a modulin from oral streptococci, requires focal adhesion kinase. J. Biol. Chem. 278: 27721-27728.
Neff, L., M. Zeisel, J. Sibilia, M. Scholler-Guinard, J. P. Klein, D. Wachsmann. 2001. NF-B and the MAP kinases/AP-1 pathways are both involved in interleukin-6 and interleukin-8 expression in fibroblast-like synoviocytes stimulated by protein I/II, a modulin from oral streptococci. Cell. Microbiol. 3: 703-712.
Alrutz, M. A., R. R. Isberg. 1998. Involvement of focal adhesion kinase in invasin-mediated uptake. Proc. Natl. Acad. Sci. USA 95: 13658-13663.(Mirjam B. Zeisel2,*, Vane)
Focal adhesion kinase (FAK) is a nonreceptor protein tyrosine kinase involved in signaling downstream of integrins, linking bacterial detection, cell entry, and initiation of proinflammatory response through MAPKs and NF-B activation. In this study, using protein I/II from Streptococcus mutans as a model activator of FAK, we investigated the potential link between FAK and TLR pathways. Using macrophages from TLR- or MyD88-deficient mice, we report that MyD88 plays a major role in FAK-dependent protein I/II-induced cytokine release. However, response to protein I/II stimulation was independent of TLR4, TLR2, and TLR6. The data suggest that there is a cross talk between FAK and MyD88 signaling pathways. Moreover, MyD88-dependent, LPS-induced IL-6 secretion by human and murine fibroblasts required the presence of FAK, confirming that MyD88 and FAK pathways are interlinked.
Introduction
Bacterial recognition implicates a wide variety of pattern recognition receptors specific for pathogen-associated microbial patterns (PAMPs), 5 including TLRs, lectins, scavenger receptors, and integrins. Among them, TLRs are critical for the development of innate immune responses (1). TLR signaling is mediated by cytoplasmic Toll/IL-1R (TIR) domain homo- or heteromeric associations with TIR-containing adaptors (2). MyD88, an adaptor protein used by all TLRs except TLR3, activates MAPKs and NF-B, leading to proinflammatory cytokine expression. MyD88-independent pathways, involving the TIR domain-containing adaptor inducing IFN- (TRIF) are used by TLR4 and TLR3, together with TRIF-related adaptor molecule in the case of TLR4. TRIF pathway activates IFN regulatory factor 3, leading to the synthesis of IFN- and -. Moreover, receptor interacting protein 1, a component of the TNFR signaling pathway, associates with the C-terminal region of TRIF and mediates TLR4 and TLR3-NF-B late activation (3). Thus, various PAMP/pattern recognition receptor interactions result in MAPK and NF-B activation and consequent expression of proinflammatory cytokines, and there is now strong evidence that many intracellular signaling molecules are shared among the activated signaling pathways. However, the kinases involved in these pathways are not yet fully elucidated.
Focal adhesion kinase (FAK), a nonreceptor protein tyrosine kinase involved in signaling downstream of integrins (4), was shown recently to trigger inflammatory responses. Protein I/II, a cell wall component from oral streptococci, binding to integrin 51 induces the production of inflammatory mediators such as IL-6 and IL-8 by human monocytes, epithelial cells, endothelial cells, and fibroblasts (5, 6). The signaling events leading to this cytokine release involve FAK, ERK1/2, and JNKs as well as AP-1-binding activity and nuclear translocation of NF-B (6, 7). FAK is an interesting candidate to link bacterial detection, initiation of proinflammatory responses, and cell entry, because FAK is involved in invasin-mediated bacterial uptake (8). Based on these observations, we investigated the possibility of a cross talk between the integrin/FAK and TLR pathways.
Using protein I/II from Streptococcus mutans as a model activator of FAK, together with macrophages from TLR- or MyD88-deficient mice, we show that MyD88 and FAK, but neither TLR2, TLR4, nor TLR6, are involved in the response to protein I/II that leads to cytokine release. This cross talk between FAK and MyD88 seems to be a general phenomenon in proinflammatory cytokine response, because LPS-induced cytokine release also depends on the presence of both FAK and MyD88, as evidenced in primary human and murine cells.
Materials and Methods
Cytokine ELISA
Cell supernatants were assayed for cytokine content using commercially available ELISA reagents for TNF-, IL-6, and IL-8 (Duoset; R&D Systems).
Nitrite measurements
Nitrite concentrations in cell supernatants were determined using the Griess reaction (3% phosphoric acid, 1% p-aminobenzenesulfonamide, 1% N-(1-napthyl)ethylenediamide) as previously described (14).
Results
Protein I/II, a ligand of integrin 51, induces FAK phosphorylation and TNF- and NO production
We showed previously that protein I/II, a ligand of integrin 51, induces cytokine production by human endothelial and epithelial cells and FLSs and that this effect is FAK dependent (5, 6). In this study, we first verified the capacity of protein I/II to stimulate phosphorylation of FAK in murine bone marrow-derived macrophages. Cell lysates were analyzed directly by blotting with specific anti-FAK (pY397) Abs. Stimulation with protein I/II resulted in an increasing amount of phosphorylated FAK (Fig. 1A), which was detectable within 5 min and remained elevated for at least 30 min. In addition, protein I/II induced a strong release of TNF- and NO by activated macrophages, reaching levels similar to those obtained after endotoxin stimulation (Fig. 1, B and C, wild-type).
The TLR2 and TLR4 independence of protein I/II activation was then confirmed in human FLSs, a cell population known to contribute to joint inflammation in rheumatoid arthritis patients. Pretreatment with Abs directed against TLR4 or TLR2 had essentially no effect on protein I/II-induced IL-8 release by FLSs (data not shown).
Taken together, these data demonstrate that TLR2, TLR4, and TLR6 are not involved in protein I/II-induced TNF- and NO release by activated murine macrophages and in protein I/II-induced IL-6 and IL-8 release by activated human FLSs.
Protein I/II-induced FAK phosphorylation is independent of MyD88
We then tested the role of MyD88 in FAK phosphorylation. Stimulation of MyD88–/– macrophages with protein I/II resulted in an increasing amount of phosphorylated FAK (Fig. 3), which was detectable within 5 min and maximal after 15 min of stimulation, comparable to what was shown in wild-type cells (Fig. 1A). Therefore, although MyD88 is involved in the response to protein I/II that leads to cytokine release, FAK phosphorylation seems to be largely independent of the presence of MyD88, suggesting that phosphorylation of FAK occurs upstream of MyD88 or that both pathways are parallel.
LPS stimulation induces FAK phosphorylation in macrophages and FLSs
Our data suggesting that MyD88 and FAK are both necessary in the signaling pathway induced by protein I/II interaction with integrin 51, we next investigated whether this cross talk between FAK and MyD88 was a general phenomenon by assessing the role of FAK in signaling pathways known to involve MyD88. Cytokine response to LPS is largely MyD88 dependent, and we examined the capacity of LPS to induce FAK phosphorylation in murine macrophages. After stimulation, cell lysates were analyzed directly by blotting with specific anti-FAK (pY397) Abs as above. Stimulation by LPS resulted in an increasing amount of phosphorylated FAK (Fig. 4A), which was detectable within 5 min and remained elevated for at least 30 min. Similar FAK phosphorylation was obtained after LPS stimulation of human FLSs (Fig. 4B). Thus, LPS induced FAK phosphorylation in murine and human primary cells.
LPS-induced IL-6 release occurs via a FAK-dependent pathway
We next asked whether FAK could contribute to the signaling events leading to cytokine release from LPS-activated cells. First, using FAK+/+ and FAK–/– primary mouse embryo fibroblasts, we showed that LPS induced IL-6 release by FAK+/+ fibroblasts, but there was essentially no IL-6 production by FAK–/– fibroblasts (p < 0.01). This suggested that FAK is involved in LPS-induced release of IL-6. To rule out the possibility that the lack of IL-6 release could be due to a nonspecific inhibition of other cellular functions, FAK+/+ fibroblasts were transfected with FAK-related nonkinase (FRNK), the C-terminal region of FAK, which is known to block FAK activation when overexpressed. FAK+/+ fibroblasts transfected with a GFP-expressing vector were used as a control. As shown in Fig. 5A, overexpression of FRNK strongly inhibited IL-6 release from LPS-activated FAK+/+ cells (p < 0.01), confirming the involvement of FAK in LPS-induced IL-6 secretion.
Taken together, our results show that MyD88 is involved in the response to protein I/II-integrin 51 interaction that leads to cytokine release through FAK phosphorylation, and this cross talk between the MyD88 and FAK pathways seems to be a general phenomenon in proinflammatory cytokine response, because MyD88-dependent, LPS-induced IL-6 secretion also depends on the presence of FAK.
Discussion
In the present study, we first demonstrated that absence of MyD88 abrogates or severely impairs TNF- and NO production by protein I/II-activated macrophages. These data suggest that MyD88, a key downstream adaptor of the TLR pathway, plays a major role in protein I/II-integrin-induced cytokine release. We showed previously, using FAK–/– and FRNK-transfected cells, that FAK is essential in IL-6 and IL-8 release by protein I/II-activated FLSs (6). Thus, we examined next how FAK and MyD88 pathways may interact. Two hypotheses were formulated, namely, protein I/II may activate MyD88 by interacting with a member of the TLR family, or alternatively, protein I/II might trigger a signaling pathway involving an intracellular cross talk between FAK and MyD88. To address these questions, we first used macrophages isolated from TLR2–/– and TLR6–/– mice insofar as these TLRs recognize various components from Gram-positive bacteria. TLR4, the receptor of LPS from Gram-negative bacteria that also interacts with fibronectin, a ligand of integrin 51, was also tested. Our results clearly show that TLR2, TLR4, and TLR6 are not essential for the stimulation of macrophages by protein I/II. Also, Abs to TLR4 and TLR2 failed to inhibit protein I/II-induced IL-8 release by human FLSs (data not shown). These data are in agreement with the observation that macrophage activation by group B streptococci requires MyD88 but is independent of TLR1, -2, -4, and -6 (17). Thus, we excluded TLR2, TLR4, and TLR6 as potential receptors for protein I/II.
We next showed that FAK plays an important role downstream of TLRs, because cytokine response to LPS is totally abrogated in FAK–/– cells. Furthermore, FAK-expressing cells transfected with the FAK inhibitor FRNK produced significantly lower amounts of IL-6 upon LPS challenge. This is in line with previous results showing that LPS induced FAK phosphorylation in a murine monocytic cell line (18). We further demonstrated that other TLR ligands may also induce FAK phosphorylation at Tyr397 and require FAK for cytokine release because comparable results have been obtained with the TLR2 ligand Pam3CSK4 (data not shown). Also, our findings extend to FAK the observation that TLR2 and TLR4 agonists induced tyrosine phosphorylation of the proline-rich tyrosine kinase 2 (Pyk2), which in turn increases tyrosine phosphorylation of paxillin, an adaptor protein involved in integrin signaling and binding to Pyk2, vinculin, and FAK, through MyD88-dependent and -independent pathways (19). However, in our experiments, MyD88 does not seem to be essential for FAK autophosphorylation because this kinase is phosphorylated in MyD88–/– macrophages activated with either LPS or protein I/II. Thus, FAK phosphorylation seems to occur upstream of MyD88, or in parallel to MyD88 recruitment, and FAK might be a key proximal effector of MyD88.
Recent studies have shown that PAMP stimulation leads to TLR recruitment to cholesterol-rich regions of the cell membrane such as lipid rafts (16). These specialized domains may contribute to the assembly of a receptor complex involving transmembrane proteins such as TLRs and integrins as well as intracellular proteins. Preliminary experiments suggest that LPS and protein I/II-induced cell activation is dependent on such lipid rafts. Indeed, disruption of raft-like domains with the cholesterol-binding drugs filipin and nystatin inhibits LPS and protein I/II-induced cytokine release, suggesting that receptor complex assembly in microdomains is essential for downstream signaling triggered by these two PAMPs (data not shown). Therefore, MyD88 and FAK may contribute to the intracytoplasmic signaling of this receptor complex.
MyD88 may be activated in at least two different ways. Ligand engagement of members of the IL-1R/TLR family induces the recruitment and phosphorylation of MyD88, but it has also been demonstrated that Fas ligation on macrophages enhances TLR4 signaling by suppressing the inhibitory interaction of Fas-associated death domain protein with MyD88 (20). One might speculate that FAK contributes to MyD88 phosphorylation, but FAK possesses neither a death domain nor a TIR domain for direct interaction with MyD88. Alternatively, FAK may participate in MyD88 activation by recruiting another kinase. In previous experiments, we addressed the involvement of Src kinases in protein I/II-induced signaling. Protein I/II did not induce phosphorylation of Src at Tyr418 (activated Src) nor mediate phosphorylation of FAK at Tyr925, a Src phosphorylation site (L. A. Neff, M. B. Zeisel, V. A. Druet, J. P. Klein, J. Sibilia, and D. Wachsmann, unpublished results). Thus, Src does not seem to play a role in protein I/II-induced signaling. It has long been known that protein tyrosine kinases are required for the induction of proinflammatory cytokines in response to LPS in macrophages. Jefferies et al. (21, 22) demonstrated Btk as a critical tyrosine kinase in LPS signaling in myeloid cells. Btk interacts with the TIR domain of TLRs as well as with downstream molecules such as MyD88, MAL, and IL-1R-associated kinase-1, but the precise nature of these interactions is not well defined. Btk is probably activated through Src kinases although other tyrosine kinases such as PI3K and Jak1 have also been shown to phosphorylate members of the Btk family (23). Because FAK has been demonstrated to be an upstream activator of Bmx/Etk in endothelial cells (24), it is conceivable that FAK might lead to the recruitment of MyD88 via the activation of a member of the Btk family of kinases and further downstream signaling to MAPKs, AP-1, and NF-B. Further studies of the involvement of additional kinase players in FAK and MyD88 pathway cross talk will be needed to complete our understanding of the molecular mechanisms linking bacterial detection, cell invasion, and proinflammatory response.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by grants from the French Ministry of Research, Pfizer, Région Alsace, Amgen, Caisse d’Assurance Maladie des Professions Libérales de Province, Wyeth Lederle, and Centre National de la Recherche Scientifique.
2 M.B.Z. and V.A.D. contributed equally to this work.
3 V.Q. and D.W. contributed equally to this work.
4 Address correspondence and reprint requests to Dr. Dominique Wachsmann, Faculté de Pharmacie, Université Louis Pasteur, 74 route du Rhin, 67400 Illkirch, France. E-mail address: wachs{at}pharma.u-strasbg.fr
5 Abbreviations used in this paper: PAMP, pathogen-associated microbial pattern; TIR, Toll/IL-1R; TRIF, TIR domain-containing adaptor inducing IFN-; FAK, focal adhesion kinase; FLS, fibroblast-like synoviocyte; Pam3CSK4, S-[2,3-bis(palmitoyloxy)-(2-RS)propyl]-N-palmitoyl-(R)-Cys-(S)-Ser-Lys4-OH, trihydrochloride; BLP, bacterial lipoprotein; FRNK, FAK-related nonkinase.
Received for publication January 5, 2005. Accepted for publication March 25, 2005.
References
Medzhitov, R.. 2001. Toll-like receptors and innate immunity. Nat. Rev. Immunol. 1: 135-145.
Takeda, K., S. Akira. 2004. Microbial recognition by Toll-like receptors. J. Dermatol. Sci. 34: 73-82.
Meylan, E., K. Burns, K. Hofmann, V. Blancheteau, F. Martinon, M. Kelliher, J. Tschopp. 2004. RIP1 is an essential mediator of Toll-like receptor 3-induced NF-B activation. Nat. Immunol. 5: 503-507.
Schlaepfer, D. D., C. R. Hauck, D. J. Sieg. 1999. Signaling through focal adhesion kinase. Prog. Biophys. Mol. Biol. 71: 435-478.[
Al-Okla, S., C. Chatenay-Rivauday, J. P. Klein, D. Wachsmann. 1999. Involvement of 51 integrins in interleukin 8 production induced by oral viridans streptococcal protein I/IIf in cultured endothelial cells. Cell. Microbiol. 1: 157-168.
Neff, L., M. Zeisel, V. Druet, K. Takeda, J. P. Klein, J. Sibilia, D. Wachsmann. 2003. ERK 1/2- and JNKs-dependent synthesis of interleukins 6 and 8 by fibroblast-like synoviocytes stimulated with protein I/II, a modulin from oral streptococci, requires focal adhesion kinase. J. Biol. Chem. 278: 27721-27728.
Neff, L., M. Zeisel, J. Sibilia, M. Scholler-Guinard, J. P. Klein, D. Wachsmann. 2001. NF-B and the MAP kinases/AP-1 pathways are both involved in interleukin-6 and interleukin-8 expression in fibroblast-like synoviocytes stimulated by protein I/II, a modulin from oral streptococci. Cell. Microbiol. 3: 703-712.
Alrutz, M. A., R. R. Isberg. 1998. Involvement of focal adhesion kinase in invasin-mediated uptake. Proc. Natl. Acad. Sci. USA 95: 13658-13663.(Mirjam B. Zeisel2,*, Vane)