- Research article
- Open Access
Flagellin acting via TLR5 is the major activator of key signaling pathways leading to NF-κB and proinflammatory gene program activation in intestinal epithelial cells
- Thomas Tallant†1,
- Amitabha Deb†1, 2,
- Niladri Kar1,
- Joseph Lupica1,
- Michael J de Veer1, 3 and
- Joseph A DiDonato1Email author
© Tallant et al; licensee BioMed Central Ltd. 2004
Received: 06 May 2004
Accepted: 23 August 2004
Published: 23 August 2004
Infection of intestinal epithelial cells by pathogenic Salmonella leads to activation of signaling cascades that ultimately initiate the proinflammatory gene program. The transcription factor NF-κB is a key regulator/activator of this gene program and is potently activated. We explored the mechanism by which Salmonella activates NF-κB during infection of cultured intestinal epithelial cells and found that flagellin produced by the bacteria and contained on them leads to NF-κB activation in all the cells; invasion of cells by the bacteria is not required to activate NF-κB.
Purified flagellin activated the mitogen activated protein kinase (MAPK), stress-activated protein kinase (SAPK) and Ikappa B kinase (IKK) signaling pathways that lead to expression of the proinflammatory gene program in a temporal fashion nearly identical to that of infection of intestinal epithelial cells by Salmonella. Flagellin expression was required for Salmonella invasion of host cells and it activated NF-κB via toll-like receptor 5 (TLR5). Surprisingly, a number of cell lines found to be unresponsive to flagellin express TLR5 and expression of exogenous TLR5 in these cells induces NF-κB activity in response to flagellin challenge although not robustly. Conversely, overexpression of dominant-negative TLR5 alleles only partially blocks NF-κB activation by flagellin. These observations are consistent with the possibility of either a very stable TLR5 signaling complex, the existence of a low abundance flagellin co-receptor or required adapter, or both.
These collective results provide the evidence that flagellin acts as the main determinant of Salmonella mediated NF-κB and proinflammatory signaling and gene activation by this flagellated pathogen. In addition, expression of the fli C gene appears to play an important role in the proper functioning of the TTSS since mutants that fail to express fli C are defective in expressing a subset of Sip proteins and fail to invade host cells. Flagellin added in trans cannot restore the ability of the fli C mutant bacteria to invade intestinal epithelial cells. Lastly, TLR5 expression in weak and non-responding cells indicates that additional factors may be required for efficient signal propagation in response to flagellin recognition.
Intestinal epithelial cells serve as a barrier between the luminal microflora and the body and as such are perfectly positioned to monitor the approach/invasion of pathogens. These intestinal epithelial cells (IECs) serve as innate immune sentinels and monitor their environment and constantly give out innate host defense instruction to local immune effector cells [1, 2]. Pathogens such as Salmonella and other enteroinvasive pathogenic bacteria such as enteroinvasive E. Coli, Shigella and Yersinia upon infection of IECs leads to the up-regulation of the expression of host genes, the products of which activate mucosal inflammatory and immune responses and alter epithelial cell functions [3–6]. Previously we and others demonstrated that IKK via NF-κB and the SAPK signaling pathways via Jun N-terminal kinase (JNK) and p38 kinase were key regulators of the up-regulation of the proinflammatory gene program [3, 7–9], with NF-κB appearing to be the most critical . Typically Salmonella infects thirty-forty percent of IECs in culture models of infection , however, we and others have found that Salmonella infection activates NF-κB DNA binding activity to levels equivalent to that of TNFα which activates NF-κB in all of the cells . Previous studies examining NF-κB activation by Salmonella in HT29 colonic intestinal epithelial cells, which serve as model colonic epithelial cells in culture, indicated that delivery of Salmonella proteins into the host cell via its type III secretion system (TTSS), such as SopE and SopE2, the bacterially encoded exchange factors for the Rho-family members Rac1 and CdC42, result in exchange factor activation, cytoskeletal rearrangements and activation of the MAPK, SAPK and NF-κB signaling pathways [7, 8, 11–15]. Recent observations that utilized Salmonella strains that were defective in invasion and delivery of invasion proteins by the TTSS but not attachment indicated that additional factors other than those delivered by the TTSS could lead to NF-κB activation . Presently it is not clear what protein(s) dictate the activation of key signaling pathways that lead to the temporal expression of the proinflammatory gene program, although the SopE proteins have been given extreme attention recently [7, 8, 15].
In searching for additional Salmonella proteins that could activate the proinflammatory gene expression program, bacterial flagellin was recently found to be such a protein [16–19] and had been shown previously to activate IL-8 expression in monocytes [19–21]. Flagellin was found to activate NF-κB in polarized epithelial cells only when flagellin was present on their basolateral surface  consistent with the idea that a cell surface receptor was present there and could recognize it. The toll-like receptors (TLRs) have been found to recognize pathogen associated molecular patterns (PAMPs) reviewed in [23–26]. TLR2 interacts with TLR1 and TLR6 to recognize bacterial lipopeptides and zymosan respectively [27, 28]. TLR4 recognizes LPS only when associated with its co-receptor MD2 and CD14 [29–32]. Recently, flagellin was demonstrated to be recognized by TLR5 and activate an innate host response [22, 33]. However, little was known or demonstrated about the endogenous levels of TLR5 in cells used in those studies and why those cells failed to respond to flagellin. Here we have identified flagellin as the primary initiator and temporal regulator of not only the major signaling pathways activated during Salmonella infection but also of key target genes of the proinflammatory gene program too. We have also found flagellin expression to be required for Salmonella bacterial invasion. Independently we found TLR5 recognizes flagellin but its signaling activity toward this PAMP is consistent with either the aid of another flagellin-recognizing co-receptor (as TLR4 utilizes for LPS) or the use of another adapter protein, perhaps similar to MyD88, that is absent or present at low levels in flagellin non-or low-responding cells.
Salmonella infection leads to a minority of cells invaded but activates NF-κB in nearly all cells
Soluble bacterial product identified as flagellin can activate NF-κB in intestinal epithelial cells
Since Salmonella sp. infection of intestinal epithelial cells in culture led to only roughly thirty percent infection but activation of NF-κB in nearly all of the cells, we anticipated that NF-κB activation was in response to host cell recognition of bacteria structural components or products produced by the bacteria and not by the invasion process. Invasion itself has been demonstrated not to be required for activation of the proinflamatory gene program as had previously been thought . To investigate this possibility sterile-filtered S. dublin culture broth left either untreated or boiled for twenty minutes was used to challenge HT29 intestinal epithelial cells and NF-κB DNA binding activity was monitored by electromobility shift assays (EMSAs) of whole cell extracts (WCE) prepared forty-five minutes after exposure [3, 35]. Potent activation of NF-κB in response to the broth under both conditions was observed indicating the activating factor was heat-stable (AD, TT and JD, personal observations) and is not LPS since HT29 cells are not responsive to LPS [3, 35].
Flagellin is required to activate NF-κB in intestinal epithelial cells
Of note is the total failure of the double flagellin gene mutants to activate NF-κB as compared to the very minor activation observed in the single Phase I flagellin fliC::Tn10 insertion mutant (next to last lanes in Fig. 4A &4B) which likely is due to the extremely limited expression of the phase II flagellin (from fljB), although the strains of Salmonella used here genetically are unable or rarely shift phases of flagellin production. These results are consistent with previous reports identifying flagellin as a potent inducer of the proinflammatory response and IL-8 production [16–19]. Since flagellin appears required for activation of the NF-κB pathway upon direct infection of intestinal epithelial cells it appeared possible that flagellin may also be the major determinant of other major mitogenic and stress activated signaling pathways activated upon pathogenic Salmonella infection of intestinal epithelial cells. Previously others and we have demonstrated that direct Salmonella infection of intestinal epithelial cells results in JNK activation  and also the activation of NF-κB via IKK . The identification of flagellin as a potent NF-κB activator is significant since SopE had previously been shown to be a pathogenic Salmonella bacteriophage encoded protein that is injected into the host cell and acts as an exchange factor for the small Rho GTPases Rac1 and CdC42 initiating cytoskeleton rearrangements and eventual activation of the MAPK, SAPK and NF-κB pathways [7, 15], while SopB is a Salmonella protein that functions as an inositol phosphate phosphatase and participates in cytoskeletal rearrangements and stimulates host cell chloride secretion .
Flagellin triggers activation of the mitogen activated protein kinase, stress activated protein kinase and IKK signaling pathways
The fliC-/fljB- double mutant Salmonella failed to invade HT29 cells compared to the wild-type Salmonella strain as determined by gentamycin protection/invasion assay (see Experimental Procedures). The flagellin fliC-/fljB- double mutant displayed a four orders of magnitude difference in its ability to invade HT29 cells (TT & JD, unpublished observations). To demonstrate this point further, we infected HT29 cells with either wild-type Salmonella or the fliC-/fljB- double mutant Salmonella (strain 134), both strains were transformed with the plasmid pFM10.1 that encodes GFP under the control of the Salmonella ssaH promoter and only functions once the bacteria has invaded the host cell [10, 34]. The wild-type Salmonella clearly was able to infect HT29 cells (GFP, Fig. 5B) while the flagellin mutant bacteria failed to invade HT29 cells as evidenced by the lack of GFP expression (Fig. 5B). To determine if flagellin is sufficient or that other bacterially produced proteins are required for invasion, we added either purified flagellin or sterile-filtered culture broths or a combination of both to HT29 cells that were challenged with the Salmonella fliC-/fljB- double mutant and assayed for invasion. Intestinal epithelial cells failed to be invaded using all tested combinations of purified flagellin and/or culture broths with the fliC-/fljB- double mutant strain (TT & JD, unpublished observations). To our knowledge there is no known direct connection between expression of flagellin genes and the effectiveness of the type III secretion system to deliver bacterially produced proteins such as SopE, SopE2 and SipA or other Sip or Sop proteins [7, 14, 15, 40, 41] that play important roles in initiating bacterial internalization. Furthermore, to evaluate the effectiveness of flagellin to stimulate p65 (RelA) nuclear localization in intestinal epithelial cells we challenged HT29 cells with purified flagellin and examined p65 (RelA) localization using indirect immunofluorescence and found p65 (RelA) nuclear localization in nearly every cell (Fig. 5B as indicated).
We wished to further examine the effect of purified flagellin and flagellin present on Salmonella on the temporal pattern of proinflammatory cytokine gene expression in intestinal epithelial cells in order to differentiate the effects of flagellin alone vs. flagellated Salmonella or non-flagellated Salmonella infection. HT29 cells were left untreated, stimulated with TNFα (10 ng/ml), or stimulated with flagellin (0.5 ug/ml), or infected with wild-type Salmonella typhimurium or the Salmonella fliC/fljB double mutant (at MOI of 50). After the indicated times after treatment or infection, HT29 cells were harvested in ice-cold PBS and the cell pellets lysed in Trizol and RNA was purified and used to prepare first-strand cDNA (see Experimental Procedures). Aliquots of the cDNA were used in semi-quantitative RT-PCR reactions using IL1α, IL-1β, IL-8, TNFα, MCP1 and β-actin gene specific primers (sequences available upon request) and the products were fractionated on ethidium bromide containing 1.2% agarose gels. Expression of the known NF-κB target genes IL-1β, IL-8, TNFα and MCP1 was increased in response to TNFα or purified flagellin exposure (Fig. 6C). Wild-type Salmonella infection also led to activation of these same genes although the expression of TNFα and MCP1 was transient in comparison and occurred immediately after infection. The Salmonella fliC-/fljB- double mutant failed to induce IL-1β, IL-8 and TNFα expression, however MCP1 expression was induced, although at lower levels than that induced by wild-type Salmonella, and also, the expression of MCP1 was not transient in nature and continued throughout the time course (9 h) (Fig. 6C). The expression level of β-actin served as an internal standard for comparison. Interestingly, IL-1α, which is not an NF-κB target gene was stimulated in response to HT29 cell challenge by all of the treatments. Obviously, the Salmonella fliC-/fljB- double mutant can activate other signaling pathways leading to IL-1α expression. We presently do not know what these signaling pathways are.
Flagellin activates NF-κB DNA binding in a MyD88-dependent manner
TLR 5 reponds to flagellin and activates NF-κB
To further determine the likelihood of TLR5 being the TLR through which flagellin activated NF-κB, we constructed dominant-negative signaling mutations by deletion of the carboxyl portion of each TLR to a conserved tryptophan in the TIR domain (see Materials and Methods). A similar mutation in the IL-1 receptor abrogates its ability to lead to NF-κB activation [51, 52]. Each DN-TLR along with a reverse cloned TLR5 (AS-TLR5) were cloned into the mammalian expression vector pCDNA3.1 (Invitrogen, Carlsbad, CA). All mutant proteins were expressed well (TT & JD unpublished observations). Each DN-TLR mammalian expression vector and empty expression vector along with 2× NF-κB Luc was transfected as previously described  into HT29 cells which respond very well to flagellin. The transfected cells were left untreated, stimulated with TNFα (10 ng/ml) or with purified flagellin (0.5 μg/ml). Reporter gene expression was observed not to be affected by DN-TLR expression in response to TNFα stimulation of transfected cells (Fig. 8A) however, only expression of either the DN-TLR5 or an antisense TLR5 construct resulted in a nearly fifty percent and twenty-five percent inhibition of flagellin-mediated reporter gene activation respectively (Fig. 8B), while DN-TLR2 also was found to mildly inhibit flagellin-mediated reporter expression. These results imply that TLR5 takes part in cell surface recognition of flagellin and initiates the signaling pathway leading to NF-κB activation. The effect of DN-TLR2 on NF-κB-dependent reporter gene activation may be non-specific since its expression also inhibited TNFα-mediated reporter activation as compared to the other DN-TLRs. DN-TLR2 may also compete for an unknown adapter protein that both TLR2 and TLR5 might share. In any event, TLR2 and TLR4 were shown by the results presented in Fig. 7 not to be required for flagellin-mediated activation of NF-κB.
Flagellin-mediated activation of NF-κB in intestinal epithelial cells leads to increased and decreased expression of a subset of TLRs
Change in TLR mRNA levels following TNFα or FliC stimulation.
TLR5 is expressed in cells that don't respond well to flagellin
This study and others [22, 33] have identified TLR5 as the likely TLR through which flagellin activates NF-κB. Previous reports made no determination on the presence or abundance of TLR5 in the cells that they used to ascertain its function [22, 33]. We wished to determine if TLR5 protein abundance was absent or greatly decreased in cells that failed to respond or responded poorly to challenge by flagellin. TLR5 abundance in a number of cell lines was examined by immunoblot analysis using a TLR5-specific antibody and compared with the ability of purified flagellin to induce NF-κB DNA binding activity of WCEs prepared from them. Intestinal epithelial cell lines T84 and HT29 were used as was the lung adenocarcinoma cell line A549, the human cervical adenocarcinoma cell line HeLa, the human embryonic kidney cell line expressing large T antigen HEK293T, and the glioblastoma cell line T98G. TLR5 protein was detected in all cell lines examined by immunoblot with TLR5-specific antibody (Fig. 9A). T84 cells exhibited the highest abundance while expression levels of the other cell lines were similar and appeared not to differ by more than two-fold (Fig. 9A). NF-κB DNA binding activity in non-stimulated, TNFα and flagellin stimulated cells was analyzed by EMSA assays of WCEs prepared from each cell type (Fig 9B). HT29 and A549 cells responded strongly to flagellin and to TNFα stimulation while HeLa, 293T and T98G cells responded poorly (HeLa, 293T) or not at all (T98G) to flagellin stimulation. The authenticity of the NF-κB DNA binding complex was determined using p65-specific antibody to supershift the NF-κB DNA:protein complex. It is of interest that some cells that express TLR5 either do not respond at all or do so very poorly. This may be due to either lack of receptor presence at the plasma membrane and intracellular localization, inactivating or detrimental mutations in the TLR5 gene in these cell lines or lack of or low abundance of a required co-receptor or adapter protein (as is the case in some cells for TLR4 and its co-receptor/adapter MD2 [30, 53, 54]). IL-1 can activate NF-κB DNA binding activity in all of the examined cell lines so it appears that the signaling apparatus downstream of MyD88 to NF-κB is intact.
Recently Muc1 a secreted and membrane bound mucin protein was shown to serve as a receptor that bound Pseudomonas aeruginosa and its flagellin, leading to activation of the MAPK pathway [55, 56] although NF-κB activity was not examined. We examined the muc1 abundance levels in HT29 (strong flagellin responder), A549 (strong flagellin responder), HeLa, 293T (both weak flagellin responders) and T98G (no flagellin response) to determine if its expression correlated with the activation profile of NF-κB and MAPK in these cells in response to flagellin . Should this be the case then muc1 might serve as a viable co-receptor for TLR5 in propagating activation signals leading to NF-κB activation. We observed that only HT29 cellular proteins gave a strong signal by immunoblot analysis using an muc1-specific antibody while muc1 was barely detectable in the other cell lines (Fig. 9C). These results suggest that muc1 does not serve the role of a TLR5 co-receptor that leads to NF-κB activation and likely plays little to no role activating MAPK pathways in A549 cells where we have observed similar temporal MAPK activation in response to flagellin exposure as we do in HT29 cells (TT and JD, unpublished results). Further examination of muc1's role in HT29 cells in regards to NF-κB and MAPK signaling using siRNA is warranted.
Intestinal epithelial cells at mucosal surfaces serve as innate immune sentinels controlling the innate host defense instruction to the immune effector cells inside the body in response to the external environment [1, 2]. Previous studies examining pathogenic Salmonella invasion of intestinal epithelial cells demonstrated activation of the proinflammatory gene program and invasion of only a minor portion of the cells . We previously demonstrated that NF-κB is as potently induced in pathogenic Salmonella .sp infected cells similar to those treated with the proinflammatory cytokines that are potent NF-κB activators such as TNFα and IL-1β and that this activity was IKK-mediated . Here we examined how bacterial invasion of only a third of the cells could give rise to NF-κB activity profiles consistent with activation of NF-κB in every cell such as the profile TNFα stimulation provides. We found that bacterial infection activates nuclear translocation of p65 (RelA) in nearly all of the intestinal epithelial cells consistent with the hypothesis that a cell surface receptor was recognizing either a soluble product that bacteria were producing, or a bacterial product on the bacteria, or both. We examined bacterial culture broths and found a bacterial product that was protein in composition and when used to challenge intestinal epithelial cells it potently activated NF-κB DNA binding activity (Fig. 2A). We further purified this protein by gel permeation and anion exchange chromatography and found the protein to be flagellin by electrospray ion trap mass spectroscopy (Fig. 2B &2C and Fig. 3). While our studies were in progress, flagellin was identified as being a potent proinflammatory mediator leading to IL-8 production and secretion [16–18]. We demonstrate in this study that flagellin appears to be exclusively responsible for activating NF-κB in intestinal epithelial cells since flagellin mutant strains do not activate NF-κB (Fig. 4) nor lead to their internalization (Fig. 5B). Furthermore, flagellin challenge of intestinal epithelial cells leads to p65 (RelA) nuclear localization in nearly all of the treated cells (Fig. 5B). Transcription factors like activator protein 1 (AP-1) and NF-κB, which are key regulators/activators of the proinflammatory gene program [57, 58] are activated by engagement of the MAPK, SAPK and IKK signaling pathways. We demonstrate that the MAPK, SAPK and IKK signaling pathways activation fails to occur in host cells by infection/exposure to Salmonella strains devoid of flagellin or products in the culture broths derived from those mutant Salmonella strains (Figs. 5A and 6B). We also demonstrate here that combined mutants of both fliC and fljB exhibit a severe lack of invasion (10-4 less than wild-type) and failure to activate stress response signaling, which has not been revealed previously. It is likely that the lack of flagellin production interferes with the functioning of the type III secretion system (TTSS) although flagellin is not known to effect expression of TTSS-required gene products. This hypothesis seems credible since supply of flagellin or bacterial culture components from wild-type Salmonella cultures in trans to the double fliC-/fljB- mutant bacteria fails to compliment their lack of infectivity in gentamycin invsion assays (TT & JD, unpublished observations and see Fig. 5B). In fact, we found the abundance of a subset of Sip and Sop proteins (SipA and SopD) released into the bacterial culture media to be drastically reduced in the flagellin mutant strains used here (TT & JD, unpublished observations). These two proteins have not previously been identified as activators of NF-κB nor are they considered as such here.
The TTSS translocates the Salmonella invasion proteins (Sips) and the SopE proteins into the host cell initiating cytoskeletal rearrangements that ultimately lead to bacterial internalization, [11, 12, 41]. In any event, it is clear that purified flagellin activates a similar cadre of proinflammatory genes as does infection of intestinal epithelial cells with wild-type flagellated Salmonella. The temporal expression pattern of these genes was found to be remarkably similar (Fig. 6C) indicating that flagellin-mediated temporal activation of the MAPK, SAPK and IKK signaling pathways can suffice for signaling pathways activated by Sips or SopE and SopE2 and largely recapitulates the temporal activation of key proinflammatory genes as does infection of intestinal epithelial cells with wild-type flagellated Salmonella.
The rapid, and potent activation of the MAPK, SAPK and IKK signaling pathways by flagellin was consistent with and indicative of the activation of a cell surface receptor. In this study and in other studies TLR5 has been demonstrated to play an integral role in the recognition of flagellin leading to activation of NF-κB and expression of the IL-8 gene (Fig. 6C) [22, 33]. Identification of TLR5 utilized transfection of TLRs 1 – 9 into cell lines which responded poorly to flagellin (this study) or not at all [22, 33] and challenging the transfectants with flagellin to identify which TLR responded to this PAMP. Previous studies that identified TLR5 as the receptor for flagellin did not examine the abundance of TLR5 in these cells or account for the lack of TLR5-mediated signaling in response to flagellin [22, 33]. We demonstrate here that cells which respond poorly (HeLa and HEK293T) or not at all (T98G) contain TLR5 in at least equivalent abundance as HT29 cells which are highly responsive to flagellin. We propose at least three possibilities to account for this discrepancy, first, this may be due to either lack of TLR5 receptor presence at the plasma membrane and intracellular localization; second, inactivating or detrimental mutations in the TLR5 gene in these cell lines; and lastly, lack of or low abundance of a required co-receptor or adapter protein required for either efficient ligand recognition and/or signaling. These possibilities are currently being investigated. We favor the last possibility since surface biotinylation experiments indicate that TLR5 is present on the cell surface in both flagellin responding cells and in non-responders mentioned above (data not shown). Invocation of the second hypothesis would require inactivating mutations be present in three different cell lines, a highly improbable outcome.
How do the findings presented here correlate with events during a "normal" Salmonella infection? We have indicated in this study that defective type III secretion system functioning leads to loss of host cell infectivity and underscores the importance of this system in the normal course of infection. In the in vivo setting, polarized epithelial cells express TLR5 on the basolateral surface  and flagellin can only reach the receptor either after either breaching the tight junction barrier by physical damage or by loosening of the junctions in response to Sips and Sops delivered into the intestinal epithelial cells by the TTSS or by delivery of flagellin across the intestinal epithelial cell by the bacteria itself [17, 59–61]. This scenario would imply the main function of the type III secretion system would be to trigger stress response signaling facilitating invasion and lead to loosening the tight junctions and result in flagellin/ flagellated bacteria to passing through the junctions and infected cells allowing access the basolateral surface and then systemic dispersion. TLR5 on the basolateral surface of the intestinal epithelial cells, in response to flagellin, could then lead to activation of NF-κB and the proinflammatory gene program and host protection. This model is consistent with activation of the proinflammatory gene program observed in response to flagellated Salmonella sp. infection in many reports too numerous to cite here and would allow the innate host defense system a fail-safe way to recognize pathogen exposure. In instances where infection of intestinal epithelial cells by naturally occurring non-flagellated Salmonella occurs, a strong proinflammatory response would not initially be presented but the Salmonella would instead lead to systemic infection as is the case in chickens with S. galinarum and S. pollorum and result in typhoid-like disease . Infection of chicken epithelial cells does not lead to proinflammatory gene expression by these non-flagellated pathogens but does when infected with S. typhimurium or S. dublin .
Argument for the existence of an additional TLR5 co-receptor/adapter being in limited abundance or absent might be in evidence from the transfection results presented in Table 1 which demonstrated that overexpression of cell surface localized FLAG-tagged TLR5 only resulted in slightly over a two-fold increase in NF-κB reporter gene expression in response to flagellin. If only TLR5 was required for activation of the signaling pathway should not a much more robust response been observed? We have also used DN-TLR5 transfections and NF-κB-dependent reporter gene assays or overexpresssion of DN-TLR5 using recombinant adenoviruses and analysis of resulting NF-κB DNA binding activity in response to flagellin to examine its effectiveness to completely inhibit TLR5-mediated flagellin activation of NF-κB. We have found it difficult to gain more than a fifty-percent reduction in either reporter gene activation or NF-κB DNA binding activity in HT29 cells (TT, AD & JD, unpublished observations). These results suggest that the resting TLR5 signaling complex may be quite stable as has recently been suggested . Should the endogenous TLR5 signaling complex be extremely stable it would therefore be expected that titration of a required pre-stimulus bound adapter or co-receptor away would be inefficient and this is what we have observed. Expression of a DN-MAL (TIRAP), a MyD88-related TLR adapter [64, 65] had little to no effect on flagellin-mediated NF-κB activity in transient transfection NF-κB reporter gene assays (TT & JD, unpublished observations). Recently, TLR5 has been shown to bind flagellin [66–68] and that this is likely a direct interaction due to failure of the human TLR5 to respond to a purified flagellin derived from a mouse-specific Salmonella strain . These observations still do not preclude the existence of a co-receptor or adapter that is critical for signal transmission. Detailed biochemical characterization of the TLR5 signaling complex will resolve this issue. Muc1, a recently described flagellin interacting membrane protein by virtue of its ability to trigger activation of the MAPK pathway in response to flagellin exposure  was considered a viable candidate for such a co-receptor but our observations suggest that it can not serve as the putative TLR5 co-receptor as it is expressed at similar levels in flagellin non-responding cell lines examined here as it is in A549 cells which respond strongly to flagellin and both cell line types express similar levels of TLR5 (Fig. 9).
In conclusion, our data clearly demonstrates that flagellin can act as the major determinant in activating key stress response signaling pathways and proinflammatory gene program expression in a temporal and qualitative fashion as observed during infection of intestinal epithelial cells by wild-type Salmonella sp. that express flagellin, a point that was not well established until this study. In addition, expression of the fli C gene appears to play an important role in the proper functioning of the TTSS since mutants that fail to express fli C are defective in expressing a subset of Sip proteins and fail to invade host cells. Flagellin added in trans cannot restore the ability of the fli C mutant bacteria to invade intestinal epithelial cells. Flagellin is "sensed" by TLR5 and in response propagates signaling pathways culminating in potent proinflammatory gene expression. Interestingly we found that TLR5 is expressed in weakly responding and also in some flagellin non-responding cells, 293T, HeLa and T98G cells respectively at levels similar to cells such as HT29 and A549 cells that respond strongly to flagellin and can be found on the cell surface, raising a strong possibility that productive TLR5 signaling may require an additional factor/adaptor other than those already known to be key in the IL-1 signaling pathway, which shares extensive similarities to the TLRs signaling pathways.
Human tumor necrosis factor alpha (TNFα) and human IL-1 were purchased from R&D Systems (Minneapolis, MN). Tris [hydroxymethyl]aminomethane (Tris) was purchased from Fisher Scientific (Fairlawn, NJ). Fetal calf serum was purchased from US Biotechnologies Inc. (Parkerford, PA). Para-nitro-phenylphosphate (PNPP) was purchased from Aldrich Chemical (Milwaukee, WI). The Polyacrylamide gel electrophoresis (PAGE) supplies: acrylamide, bis-acrylamide, sodium dodecyl sulfate (SDS), TEMED, and ammonium persulfate were purchased from Bio-Rad Laboratories (Hercules, CA). Dulbecco's modified essential medium (DMEM), DMEM:F12, phosphate buffered saline (PBS), glutamine, penicillin G, streptomycin, amphotericin B, and Grace's Insect medium were purchased from Invitrogen (Carlsbad, CA). Luria Broth (LB) was purchased from Becton Dickson and Co (Sparks, MD). The protease inhibitors: aprotinin, bestatin, leupeptin, pepstatin A, and phenylmethylsulfonyl fluoride (PMSF) were purchased from Cal Biochem (La Jolla, CA). Protease inhibitor cocktail contained 10 μg/ml aprotinin, 2.5 μg/ml leupeptin, 8.3 μg/ml bestatin, and 1.7 μg/ml pepstatin A. Phorbol 12-myristate 13 acetate (PMA), N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] (Hepes), anisomycin, and 2-[N-morpholino]ethanesulfonic acid (MES) were purchased from Sigma Chemical (St. Louis, MO). All other reagents were purchased from Sigma Chemical or Fisher Scientific unless stated otherwise.
HT29 human intestinal (colorectal adenocarcinoma) epithelial cells (ATCC HTB-38), HeLa cervical epithelial adenocarcoma cells (ATCC CCL-2), 293T kidney cells (CRL-11268), A549 lung carcinoma cells (ATCC-185), and T98G glioblastoma cells (ATCC CRL-1690) were cultured in DMEM with 2 mM glutamine, 10% Fetal Calf Serum, 100 Units/ml Penicillin G, and 100 μg/ml Streptomycin at 37°C in a humidified 5% CO2 atmosphere. T84 colorectal carcinoma cells (ATCC CCL-248) were cultured in DMEM:F12 with 2 mM glutamine, 5% Fetal Calf Serum, 100 Units/ml Penicillin G, and 100 μg/ml Streptomycin at 37°C in a humidified 5% CO2 atmosphere. H5 insect cells (Invitrogen) were cultured in Grace's medium with 2 mM glutamine, 10% Fetal Calf Serum, 100 Units/ml Penicillin G, 100 μg/ml Streptomycin, and 0.25 μg/ml amphotericin B at 28°C. MyD88-/- & TLR2-/-/TLR4-/- double knockout cells were obtained from Shizuo Akira and Osamu Takeuchi (Univ. of Osaka, Japan) and grown in DMEM with 2 mM glutamine, 10% Fetal Calf Serum, 100 Units/ml Penicillin G, and 100 μg/ml Streptomycin at 37°C in a humidified 5% CO2 atmosphere.
Salmonella typhimurium strain SJW1103 (FliC, phase 1 flagellin, stabilized)  is a wild-type Salmonella typhimurium and can only express the Phase I fliC flagellin, SJW86 (SJW1103 FliC::TN10), and SJW134 (SJW1103 FliC and FljB deletions) were obtained from Robert Macnab (Yale Univ., Conn) and have been described . Salmonella serovar dublin strain 2229, strain SE1 (2229 SopE mutant), strain SB2 (2229 SopB mutant), and SE1SB2 (2229 SopE and SopB mutant) were obtained from Edward Galyov (Compton Laboratory, Berkshire, UK) and have been described [14, 15]. Salmonella strains for stimulation were grown in LB at 37°C without agitation for 16 hours, centrifuged at 6,000 × g for 1 minute, gently washed with PBS, and gently suspended in DMEM to maintain cells with attached flagella.
Plasmid pFM10.1 (ampicillin resistance), encodes a green fluorescent protein (GFP) expressed after the Salmonella host is internalized by mammalian cells, obtained from Stanley Falkow (Stanford Univ., Stanford, CA) [10, 34] and was transformed into strains SJW1103 and SJW134 by electroporation. Strains containing pFM10.1 were designated SJW1103G and SJW134G.
Preparation and analysis of Salmonella cell free culture supernatant
Native flagellin was harvested from S. dublin 2229 or S. typhimurium SJW1103. Starter cultures were grown in Luria broth (LB) for 18 hours at 37°C with aeration, diluted 1:5000 in fresh LB, and grown for 12 hours under the same conditions. All subsequent procedures were performed at 4°C. Cells were removed from the medium by centrifugation at 10,000 × g for 5 min and discarded. The supernatant containing flagellin was filtered through a 0.8 micron filter (Millipore, Bedford, MA) to remove residual cells. Supernatant was concentrated 100 fold using an Amicon 100 kiloDalton (kDa) cutoff membrane (Millipore). Initial studies used concentrated culture supernatant from S. dublin strain 2229 that was treated with DNase, RNase, Protease K, boiled for 20 min or 100 mM DTT at 37°C for 2 hours and used for stimulation of cultured cells.
Concentrated S. typhimurium 1103 bacterial culture supernatant was washed 4 times by 1:10 dilution with 50 mM MES, pH 6.0, 50 mM NaCl and re-concentrated. Material not retained by the 100 kDa membrane was discarded. Washed culture supernatant was fractionated by gel permeation or anion exchange chromatography for analysis. For long-term storage, washed culture supernatant was supplemented with protease cocktail and stored at -20°C.
Fractionation by gel permeation chromatography was performed with a Superose 12HR column (Pharmacia) on a Bio-Logic system (Bio-Rad). One-half mililiter of 100× washed supernatant (equivalent of 50 ml original culture supernatant) was separated on the column at 0.4 ml/minute in 50 mM Hepes, pH 7.4, 200 mM NaCl. Fractions (0.5 ml) were collected, and 50 μl was fractionated by SDS-PAGE and stained with Bio-Safe Coomassie (Bio-Rad). Thirty microliters of each fraction was used for stimulation of HT29 cells (60 mm dishes) for 45 min and NF-κB DNA binding activity in the resulting whole cell extracts extracts were assayed by EMSA. The column was standardized with catalase (232 kDa), aldolase (158 kDa), abumin (67 kDa), ovalbumin (43 kDa), and Chymotripsinogen A (25 kDa), all obtained from Amersham-Pharmacia.
Fractionation by anion exchange chromatography was performed with Poros HQ matrix (2 ml column, PerSeptive Biosystems, Farmingham, MA) on a Bio-Logic system. Five mililiters of 100× washed supernatant (equivalent of 500 ml original culture supernatant) was separated at 1 ml/minute in 50 mM Hepes, pH 7.4, and a NaCl gradient from 50–500 mM. Fractions were collected and 5 μl of each fraction was examined by 10% SDS-PAGE. Proteins were fractionated on duplicate 10% SDS-PAGE precast gels (BioRad). One gel was stained with Bio-Safe Coomassie (Bio-Rad) and the protein bands were isolated for Mass Spectroscopy analysis (CCF Mass spectroscopy core facility) from the other identical non-stained gel, by electro-elution with a whole gel eluter (Bio-Rad) and SDS was removed with SDS-Out (Pierce, Rockland, IL) per the manufacturers directions. Proteins isolated from bands B1 to B6 were acetone precipitated by addition of 20 μg Aprotinin and 5 μg of BSA to each eluted fraction, ice-cold acetone (-20°C) was added to 80%, mixed well and precipitated overnight at -20°C. Proteins were pelleted by centrifugation at 14,000 × g in the cold for 30 min, acetone/liquid was removed and the pellets washed 2× with 1 ml acetone (-20°C). After removal of the acetone, protein pellets were air dried and then resuspended and denatured in 5 μl of 6 M guanidinium hydrochloride (Gu-HCl) at room temperature for 30 min. Resuspended proteins were two-fold serially diluted in DMEM to a final Gu-HCl concentration of 55 mM to renature the proteins. Two hundred fifty microliters of individual renatured proteins/DMEM were added per ml to HT29 cells (60 mm dishes) and whole cell extracts were prepared 45 min after stimulation and were assayed for NF-κB DNA binding activity by EMSA.
Purification of flagellin (purified flagellin)
The washed and concentrated culture supernatant from S. typhimurium 1103 containing flagellin was boiled for 20 minutes and precipitants removed by centrifugation at 15,000 × g. The supernatant containing flagellin was diluted 1:2 with 50 mM MES, pH 6.0, 50 mM NaCl and mixed with 2 ml Poros SP cation exchange matrix (PerSeptive Biosystems) per 1 liter of original culture. The Poros SP matrix was prepared as a 50% slurry and equilibrated with 50 mM MES, pH 6.0. The flagellin preparation and matrix were mixed on a roller at 12 to 14 RPM for 2 hours. The matrix along with bound contaminants was removed by filtration through a 0.85 micron filter and discarded, flagellin failed to bind to the cation exchange matrix at pH 6.0 and eluted in the flowthrough and was collected.
The pH of the flowthrough was adjusted by five-fold dilution of the sample with 50 mM Hepes, pH 7.8, 50 mM NaCl, and loaded onto a Poros HQ anion exchange column (2 ml column, PerSeptive Biosystems) equilibrated with 50 mM Hepes, pH 7.4, 50 mM NaCl. The column was washed with 2 volumes 50 mM Hepes, pH, 7.4, 50 mM NaCl, and eluted with a 10 column volume linear gradient of 50–500 mM NaCl in 50 mM Hepes, pH 7.4. Flagellin eluted from the column between 200–275 mM NaCl. Fractions containing flagellin were pooled and concentrated. The preparation was determined to be pure by electrophoresis of 5 μg protein by SDS-PAGE and stained with Bio-Safe Coomassie (Bio-Rad). Samples were stored at -80°C in 50 mM Hepes, pH 7.4, approx 225 mM NaCl, 10% glycerol and protease cocktail. A 4 liter preparation of culture supernatant yielded 2 mg purified flagellin.
In-gel tryptic digestion and protein identification by LC-MS
Gels were fixed and stained (Bio-Safe Blue, BioRad). All of the following procedures were performed by the CCF Mass spectroscopy core facility. Excised gel bands were reduced (100 mM DTT), and alkylated (100 mM iodoacetamide). Proteins in the gel bands were digested with modified trypsin (Promega, 20 μg/mL) with an overnight incubation at 37°C. Tryptic peptides were extracted from the gel with 50% acetonitrile, 0.1% acetic acid, concentrated in a SpeedVac (Thermo Savant) to remove acetonitrile, and reconstituted to 20 uL with 0.1% acetic acid. Extracted peptides were subjected to reversed phase (50 uM ID packed with Phenomenex Jupiter C18, 6 cm capillary column) liquid chromatography (2%–70% solvent B; Solvent A, 50 mM acetic acid, aqueous, Solvent B acetonitrile), coupled to a Finnigan LCQ DECA ion trap mass spectrometer for peptide sequencing, as described .
Preparation of GST-IκBa1-54 and GST-cJUN1-79 kinase substrates
IκBα amino acids 1 to 54 fused to GST or cJUN amino acids 1–79 fused to GST were prepared as previously described [37–39] and stored in kinase buffer (20 mM Hepes, pH 7.6, 10 MM MgCl2, 10 mM NaCl, 2 mM beta-glycerophosphate, 10 mM PNPP).
Preparation of cells for microscopy
HT29 cells for microscopic examination were grown in 6 well plates on sterile cover slips to a density of 50–75%. Cells were stimulated as described above. After stimulation, cover slips with HT29 cells were washed 2 times with ice cold PBS and fixed with 4% w/v formalin at room temperature for 20 minutes. Cells were washed 4 times with PBS prior to mounting for visualization of Salmonella invasion. Cover slips were mounted with Vectashield mounting medium with DAPI (Vector Laboratories, Burlingame, CA), and cover slips sealed to slides.
Cells for antibody staining were treated with absolute methanol for 20 minutes following formalin fixation, then washed 3 times with PBS supplemented with 0.1% BSA (PBSB) and used directly or stored in the cold after azide was added to 0.02%. For p65(RelA) localization, cells on coverslips were blocked for 1 h at 37°C with PBS supplemented with 1% BSA. The PBSB was removed, washed once with PBSB and coverslips were placed cell-side down onto 150 μl of p65 antibody (Zymed, South San Francisco, CA) diluted 1:1500 in PBSB on a square of parafilm and placed in a humidified chamber at 37°C for 1.5 h. Coverslips were removed and placed cell-side up in 6-well dishes and washed 3 × 5 min with PBSB. Coverslips were then removed and placed cell-side down onto 150 μl of FITC-labeled donkey anti-rabbit secondary antibody (Jackson Immunoresearch Laboratories, West Grove, PA) (1:300 in PBSB) on a square of parafilm and placed in a humidified chamber at 37°C for 1.5 h. Coverslips were removed and placed cell-side up in 6-well dishes and washed 5 × 5 min with PBSB, removed and placed cell-side down onto slides mounted with Vectashield (Vector Laboratories, Burlingame, CA) with DAPI and then sealed. NF-κB localization was determined by indirect immunofluorescence. Samples were observed on a Leica DMR upright microscope (Leica Microsystems Inc., Heidelberg, Germany) at 400× with oil immersion and equipped with FITC and UV filters. Images were collected with a MicroMax RS camera (Princeton Instruments Inc., Princeton, NJ), and Image Pro plus, version 4.5, software (Media Cybermetics Inc., Carlsbad, CA). Color enhancements were performed with Image Pro plus software. Visible light plus color overlays for Fig. 1 and Fig. 5B were performed with MetaMorph Software (Universal Imaging Corp., Downington, PA).
Bacterial infection and cell stimulation
Mouse embryo fibroblasts (MEFs) or HT29 cells were grown in DMEM as above to a density of 90% prior to stimulation. All cells were washed with warm PBS and supplemented with DMEM without serum or antibiotics in preparation for stimulation. Cells were stimulated with; 10 ng/ml TNFα, 1 μg/ml flagellin unless specified otherwise, 20 μg/ml Anisomycin, 12.5 ng/ml PMA, or 108 Salmonella/ml at 37°C for desired times and extracts prepared as below. Cells harvested beyond one hour were washed with warm PBS and supplemented with warm DMEM, 2 mM glutamine, and 200 ug/ml gentamycin after 1 hour and returned to 37°C until extract preparation desired.
Whole cell extract preparation
Cells were washed with ice-cold PBS and all subsequent steps carried out at 4°C or on ice. Cells were scraped from the dish in ice-cold PBS, and collected by centrifugation at 1000 × g for 1 minute. Cells were lysed by suspension in 50 mM Tris-HCl, pH 7.6, 400 mM NaCl, 25 mM beta-glycerol phosphate, 25 mM NaF, 10 mM PNPP, 10 % glycerol, 0.5 mM sodium orthovanadate, 0.5% nonidet-40 (NP-40), 5 mM benzamidine, 2.5 mM metabisulfite, 1 mM PMSF, 1 mM DTT and protease inhibitor cocktail as described .
Electromobility shift assays (EMSA)
NF-κB DNA binding assays were carried out as previously described [3, 35, 38]. Anti-p65 antibody (Zymed, South San Francisco), anti-p50 antibody (Santa Cruz Biotechnologies, Santa Cruz, CA), and anti-STAT3 antibody (Santa Cruz) were used for EMSA supershifts.
HT29 cells, 90–95% confluent in 35 mm round dishes, were prepared for stimulation as above and treated with a 1 ml suspension of Salmonella SJW1103 or SJW134 or left untreated in triplicate as above. After one hour, HT29 cells were washed 4× with warm PBS, supplemented with warm DMEM, 2 mM glutamine, and 200 μg/ml gentamycin, and incubated at 37°C for 4 hours. Cells were then harvested as above and lysed by suspension in 1 ml sterile distilled water. Ten-fold serial dilutions were prepared in PBS and 100 μl of each dilution was plated on LB agar plates and grown at 37°C for 20 hours. Colonies were counted and averaged.
Whole cell extracts (250 μg) were supplemented with 150 μl of Buffer A (20 mM Hepes, pH 7.9, 20 mM beta-glycerophosphate, 10 mM NaF, 0.1 mM orthovanadate, 5 mM PNPP, 10 mM 2-mercaptoethanol, 0.5 mM PMSF, and protease inhibitor cocktail), and immuno precipitation kinase assays carried out as described  using either IKKα monoclonal antibody (PharMingen – Becton Dickson), anti-JNK1 (Santa Cruz Biotechnologies, Santa Cruz, CA), or anti-hemagglutinin (HA) epitope antibody (Covence Antibodies, Princeton, NJ) as indicated. Protein G immunopellets were collected by centrifugation at 500 × g for 30 sec, washed 3 times with Buffer B (Buffer A plus 250 mM NaCl), and one time with Buffer C (Buffer A plus 50 mM NaCl and 10 mM MgCl2). Immunopellets were resuspended in 30 μl Kinase buffer with 0.1 mM orthovanadate, 50 μM "cold" ATP, 5 μCi γ-32P-ATP, 2 mM DTT, and 2 μg of soluble GST-IκBα1–54 or GST-cJUN1-79, and incubated at 30°C for 30 minutes. Reactions were stopped by the addition of 15 μl 4× SDS-PAGE loading buffer, heated at 95°C for 5 minutes, and resolved on 10% SDS-PAGE gels by standard procedures. Gels were rinsed, stained with Bio-Safe Coomassie (Bio-Rad) to visualize protein bands, rinsed, photographed then dried and exposed to Kodak X-OMAT AR film (Eastman Kodak Co., Rochester, NY) to detect substrate phosphorylation.
Protein samples (40 μg) were resolved by SDS-PAGE on a 10% acrylamide gels by standard procedures, and proteins transferred to PVDF membrane (Millipore) and probed with antibodies as described . Membranes were washed 3× briefly with TBST, incubated with a 1:1000 dilution (1:800 for anti-TLR5) of the primary antibody in TBST, 1% non-fat milk for 1 hour, washed 3 × 5 min with TBST, and then incubated with a 1:2000 dilution of the appropriate HRP-conjugated secondary antibody in TBST, 0.5% non-fat milk for 1 hour. Primary antibodies used were: anti-IKKα/β (H-470, Santa Cruz), anti-JNK1, anti-ERK2 (K-23, Santa Cruz), anti-phospho-ERK (E-4, Santa Cruz), anti-p38MAPK (Cell Signaling Technologies, Beverly, MA), anti-phosopho-p38MAPK (Cell Signaling), anti-TLR5 (H-127, Santa Cruz), anti-muc1 (H-295, Santa Cruz) and anti-actin (C-11, Santa Cruz). Secondary antibodies used were: anti-mouse IgG HRP conjugate (Amersham-Pharmacia), anti-rabbit IgG HRP conjugate (Amersham-Pharmacia), anti-goat IgG-HRP conjugate (Santa Cruz). HRP activity was detected by ECL (Amersham-Pharmacia) as per manufacturers instructions, on Kodak X-OMAT AR film.
Construction of dominant-negative TLRs
All DN-TLRs were constructed using PCR. The universal 5' primer consisted of a 5'KPN I restriction site followed by sequences encoding the kozak sequence, translational start site, and preprotrypsin leader sequence of pCMV-1 (Sigma) that all the wild-type TLRs were initially cloned into. The 3' anti-sense (AS) primers were human TLRgene-specific primers (sequences available upon request) that created a stop codon immediately after a conserved tryptophan in Box 9 of the TLR TIR homology domain according to Bazan , thus creating carboxy terminus deletions. The 5' end of the AS primer contained a number of convenient restriction sites to allow directional cloning. PCR was performed with turbo-Pfu polymerase (Stratagene, La Jolla, CA) using standard procedures on individual wild-type TLR pCMV-1 plasmid DNAs (5 ng each, kind gifts of R. Medzhitov, Yale Univ. and R. Ulevitch, TSRI) [48, 49] with the 150 ng each of the universal 5' sense primer and individual gene-specific TLR 3' primers. PCR products were cleaned-up with PCR cleanup kit (Qiagen, Germany) digested with appropriate restriction enzymes, gel purified and then ligated into the mammalian expression vector pCDNA3.1 (Invitrogen). Positive clones were sequenced to verify the mutations and tested for expression in transient expression assays and detected on immunoblots by probing with anti-FLAG M2 monoclonal antibody (Sigma). All wild-type and DN-TLR alleles are amino terminus FLAG epitope-tagged.
HT29 cells were transfected with Lipofectamine Plus (Invitrogen) as previously described . In transfections monitoring reporter gene expression, transfections were performed at least three times in 6 well dishes in triplicate with the total DNA mass kept constant at 4 μg (2 μg effector plasmid DNA, 100 ng 2× NF-κB Luc reporter gene, 50 ng pRL-TK, a thymidine kinase promoter driven Renilla luciferase normalization reporter and 1.85 μg pCDNA3.1 plasmid DNA as bulk filler DNA) and fire-fly luciferase expression was normalized to Renilla luciferase expression using the dual-luciferase assay (Promega, Madison, WI). Fold inductions were calculated and values between experiments did not vary more than 15%, a representative experiment is presented. Transfection of 293T cells was performed with lipofectamine 2000 (Invitrogen) in 6-well dishes in triplicate as per the manufacture's protocol. TLR expression plasmids were added at 2 μg/well, and NF-κB and normalization control plasmids were as above with HT29 cells and pCDNA3.1 plasmid DNA as bulk filler DNA to a final DNA mass of 4 μg/well. Fold inductions were calculated and values between experiments (N of 3) did not vary more than 10%, a representative experiment is presented.
Real Reverse Transcription and Real Time PCR (RT2PCR)
Cells (N = 3) were stimulated 3 hours at 37°C with TNFα or FliC or left untreated and harvested for total RNA isolation. Total cellular RNA was extracted from cells with Trizol reagent (Invitrogen)  and reverse transcribed with ReactionReady first strand cDNA synthesis kit (SuperArray Bioscience Corp., Fredrick, MD). RNA (2.5 ug per 20 ul reaction) was reverse transcribed using random primers and Moloney murine leukemia virus reverse transcriptase per manufacturer specified conditions. Controls without reverse transcriptase (minus RT) was also generated for each RNA sample. RT2PCR was performed with an iCycler (Bio-Rad) to quantify TLR1 through TLR10 mRNA, 18S rRNA, and GAPDH mRNA. RT2PCR (25 ul reaction volume) was performed with the appropriate primers (SuperArray) per manufacturers instructions in triplicate with HotStart Taq DNA polymerase (SuperArray) at 95°C for 15 min to activate Taq and amplified for 40 cycles (95°C, 30 sec, 55°C, 30 sec, 72°C, 30 sec). RT2PCR was performed on the minus RT controls with TLR5 primers to detect DNA contamination. Real-time PCR analysis was performed using SYBR-green (Perkin-Elmer) according to manufacture's instructions with the specific primer pairs indicated above and primer pairs for 18S ribosomal RNA as reference RNA (Classic 18S primer pairs – Ambion Inc). Cycle time (Ct) was measured using the iCycler™ and its associated software (Bio-Rad). Relative transcript quantities were calculated by the ΔΔCt method using 18S ribosomal RNA as a reference amplified from samples using the Classic 18S primer pairs from Ambion, Inc (Austin, TX). Normalized samples were then expressed relative to the average ΔCt value for untreated controls to obtain relative fold-change in expression levels. Fold change in mRNA expression was expressed as 2ΔΔCt. ΔCt is the difference in threshold cycles for the TLR mRNAs and 18S rRNA. ΔΔCt is the difference between ΔCt non-simulated control and ΔCt stimulated sample. Values for fold-induction varied less than 5% among replicates.
The abbreviations used are
fetal bovine serum
interlukin-1, SDS, sodium dodecyl sulfate
polyacrylamide gel electrophoresis
electromobility shift assay
tumor necrosis factor α
nuclear factor kappa B
Ikappa B kinase
polymerase chain assay
reverse transcription polymerase chain assay
mitogen activated protein kinase
stress-activated protein kinase
extracellular regulated kinase
Jun N-terminal kinase
mouse embryo fibroblast
whole cell extract
intestinal epithelial cell
macrophage chemoattractant protein 1
type III secretion system
Salmonella invasion protein
phorbol 12-myristate 13 acetate
para nitrophenyl phosphate
normal rabbit serum
We would like to thank Drs.E.Galyor (Compton Laboratory, Berkshire (UK)), Drs. R.M. Macnab (Yale University, New Haven, Conn), S. Mizel (Wake Forest University, Winston-Salem, NC), and M. Kagnoff (UCSD, La Jolla, CA) for bacterial strains. We would also like to thank S. Akira and O. Takeuchi (Osaka University, Osaka, JA) for the gift of MyD88 and TLR 2/4 double knockout cell lines. We also would like to thank S. Falkow (Stanford University, Stanford, CA), M. Karin (UCSD, La Jolla, CA), R. Medzhitov (Yale University, New Haven, Conn) and R. Ulevitch (TSRI, La Jolla, CA) and V. Dixit (then at Univ. of Michigan, Ann Arbor, MI) for gifts of plasmids. We would also like to thank B. Williams (CCF, Cleveland, OH) for thoughtful discussions and support on this project. This work was supported in part by grants from the National Institutes of Health, CA84406 (to J.A.D.) and the U.S. Army, DAMD 17-01-C-0065 (to B. Williams).
- Kagnoff MF, Eckmann L: Epithelial cells as sensors for microbial infection. J Clin Invest. 1997, 100: 6-10.PubMed CentralView ArticlePubMedGoogle Scholar
- Eckmann L, Kagnoff MF, Fierer J: Intestinal epithelial cells as watchdogs for the natural immune system. Trends Microbiol. 1995, 3: 118-120. 10.1016/S0966-842X(00)88894-0.View ArticlePubMedGoogle Scholar
- Elewaut D, DiDonato JA, Kim JM, Truong F, Eckmann L, Kagnoff MF: NF-kappa B is a central regulator of the intestinal epithelial cell innate immune response induced by infection with enteroinvasive bacteria. J Immunol. 1999, 163: 1457-1466.PubMedGoogle Scholar
- Witthoft T, Eckmann L, Kim JM, Kagnoff MF: Enteroinvasive bacteria directly activate expression of iNOS and NO production in human colon epithelial cells. Am J Physiol. 1998, 275: G564-G571.PubMedGoogle Scholar
- Eckmann L, Stenson WF, Savidge TC, Lowe DC, Barrett KE, Fierer J, Smith JR, Kagnoff MF: Role of intestinal epithelial cells in the host secretory response to infection by invasive bacteria. Bacterial entry induces epithelial prostaglandin h synthase-2 expression and prostaglandin E2 and F2 alpha production. J Clin Invest. 1997, 100: 296-309.PubMed CentralView ArticlePubMedGoogle Scholar
- Jung HC, Eckmann L, Yang SK, Panja A, Fierer J, Morzycka-Wroblewska E, Kagnoff MF: A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion. J Clin Invest. 1995, 95: 55-65.PubMed CentralView ArticlePubMedGoogle Scholar
- Hardt WD, Chen LM, Schuebel KE, Bustelo XR, Galan JE: S. typhimurium encodes an activator of Rho GTPases that induces membrane ruffling and nuclear responses in host cells. Cell. 1998, 93: 815-826. 10.1016/S0092-8674(00)81442-7.View ArticlePubMedGoogle Scholar
- Hobbie S, Chen LM, Davis RJ, Galan JE: Involvement of mitogen-activated protein kinase pathways in the nuclear responses and cytokine production induced by Salmonella typhimurium in cultured intestinal epithelial cells. J Immunol. 1997, 159: 5550-5559.PubMedGoogle Scholar
- Chen LM, Hobbie S, Galan JE: Requirement of CDC42 for Salmonella-induced cytoskeletal and nuclear responses. Science. 1996, 274: 2115-2118. 10.1126/science.274.5295.2115.View ArticlePubMedGoogle Scholar
- Valdivia RH, Hromockyj AE, Monack D, Ramakrishnan L, Falkow S: Applications for green fluorescent protein (GFP) in the study of host-pathogen interactions. Gene. 1996, 173: 47-52. 10.1016/0378-1119(95)00706-7.View ArticlePubMedGoogle Scholar
- Galan JE: Salmonella interactions with host cells: type III secretion at work. Annu Rev Cell Dev Biol. 2001, 17: 53-86. 10.1146/annurev.cellbio.17.1.53.View ArticlePubMedGoogle Scholar
- Galan JE, Collmer A: Type III secretion machines: bacterial devices for protein delivery into host cells. Science. 1999, 284: 1322-1328. 10.1126/science.284.5418.1322.View ArticlePubMedGoogle Scholar
- Kubori T, Galan JE: Salmonella type III secretion-associated protein InvE controls translocation of effector proteins into host cells. J Bacteriol. 2002, 184: 4699-4708. 10.1128/JB.184.17.4699-4708.2002.PubMed CentralView ArticlePubMedGoogle Scholar
- Bakshi CS, Singh VP, Wood MW, Jones PW, Wallis TS, Galyov EE: Identification of SopE2, a Salmonella secreted protein which is highly homologous to SopE and involved in bacterial invasion of epithelial cells. J Bacteriol. 2000, 182: 2341-2344. 10.1128/JB.182.8.2341-2344.2000.PubMed CentralView ArticlePubMedGoogle Scholar
- Wood MW, Rosqvist R, Mullan PB, Edwards MH, Galyov EE: SopE, a secreted protein of Salmonella dublin, is translocated into the target eukaryotic cell via a sip-dependent mechanism and promotes bacterial entry. Mol Microbiol. 1996, 22: 327-338. 10.1046/j.1365-2958.1996.00116.x.View ArticlePubMedGoogle Scholar
- Eaves-Pyles T, Szabo C, Salzman AL: Bacterial invasion is not required for activation of NF-kappaB in enterocytes. Infect Immun. 1999, 67: 800-804.PubMed CentralPubMedGoogle Scholar
- Gewirtz AT, Simon PO, Schmitt CK, Taylor LJ, Hagedorn CH, O'Brien AD, Neish AS, Madara JL: Salmonella typhimurium translocates flagellin across intestinal epithelia, inducing a proinflammatory response. J Clin Invest. 2001, 107: 99-109.PubMed CentralView ArticlePubMedGoogle Scholar
- Gewirtz AT, Rao AS, Simon PO, Merlin D, Carnes D, Madara JL, Neish AS: Salmonella typhimurium induces epithelial IL-8 expression via Ca(2+)-mediated activation of the NF-kappaB pathway. J Clin Invest. 2000, 105: 79-92.PubMed CentralView ArticlePubMedGoogle Scholar
- McDermott PF, Ciacci-Woolwine F, Snipes JA, Mizel SB: High-affinity interaction between gram-negative flagellin and a cell surface polypeptide results in human monocyte activation. Infect Immun. 2000, 68: 5525-5529. 10.1128/IAI.68.10.5525-5529.2000.PubMed CentralView ArticlePubMedGoogle Scholar
- Ciacci-Woolwine F, Blomfield IC, Richardson SH, Mizel SB: Salmonella flagellin induces tumor necrosis factor alpha in a human promonocytic cell line. Infect Immun. 1998, 66: 1127-1134.PubMed CentralPubMedGoogle Scholar
- Moors MA, Li L, Mizel SB: Activation of interleukin-1 receptor-associated kinase by gram-negative flagellin. Infect Immun. 2001, 69: 4424-4429. 10.1128/IAI.69.7.4424-4429.2001.PubMed CentralView ArticlePubMedGoogle Scholar
- Gewirtz AT, Navas TA, Lyons S, Godowski PJ, Madara JL: Cutting edge: bacterial flagellin activates basolaterally expressed TLR5 to induce epithelial proinflammatory gene expression. J Immunol. 2001, 167: 1882-1885.View ArticlePubMedGoogle Scholar
- Akira S: Toll-like receptors and innate immunity. Adv Immunol. 2001, 78: 1-56. 10.1016/S0065-2776(01)78001-7.View ArticlePubMedGoogle Scholar
- Akira S, Hemmi H: Recognition of pathogen-associated molecular patterns by TLR family. Immunol Lett. 2003, 85: 85-95. 10.1016/S0165-2478(02)00228-6.View ArticlePubMedGoogle Scholar
- Medzhitov R, Janeway C: The Toll receptor family and microbial recognition. Trends Microbiol. 2000, 8: 452-456. 10.1016/S0966-842X(00)01845-X.View ArticlePubMedGoogle Scholar
- O'Neill L: Specificity in the innate response: pathogen recognition by Toll-like receptor combinations. Trends Immunol. 2001, 22: 70-10.1016/S1471-4906(00)01842-1.View ArticlePubMedGoogle Scholar
- Hajjar AM, O'Mahony DS, Ozinsky A, Underhill DM, Aderem A, Klebanoff SJ, Wilson CB: Cutting edge: functional interactions between toll-like receptor (TLR) 2 and TLR1 or TLR6 in response to phenol-soluble modulin. J Immunol. 2001, 166: 15-19.View ArticlePubMedGoogle Scholar
- Ozinsky A, Underhill DM, Fontenot JD, Hajjar AM, Smith KD, Wilson CB, Schroeder L, Aderem A: The repertoire for pattern recognition of pathogens by the innate immune system is defined by cooperation between toll-like receptors. Proc Natl Acad Sci U S A. 2000, 97: 13766-13771. 10.1073/pnas.250476497.PubMed CentralView ArticlePubMedGoogle Scholar
- Kawasaki K, Akashi S, Shimazu R, Yoshida T, Miyake K, Nishijima M: Involvement of TLR4/MD-2 complex in species-specific lipopolysaccharide-mimetic signal transduction by Taxol. J Endotoxin Res. 2001, 7: 232-236. 10.1179/096805101101532738.View ArticlePubMedGoogle Scholar
- Nagai Y, Akashi S, Nagafuku M, Ogata M, Iwakura Y, Akira S, Kitamura T, Kosugi A, Kimoto M, Miyake K: Essential role of MD-2 in LPS responsiveness and TLR4 distribution. Nat Immunol. 2002, 3: 667-672.PubMedGoogle Scholar
- Backhed F, Normark S, Richter-Dahlfors A: TLR4-dependent lipopolysaccharide signalling in epithelial cells is independent of extracellular protease activity. Cell Microbiol. 2002, 4: 297-303. 10.1046/j.1462-5822.2002.00190.x.View ArticlePubMedGoogle Scholar
- da Silva CJ, Soldau K, Christen U, Tobias PS, Ulevitch RJ: Lipopolysaccharide is in close proximity to each of the proteins in its membrane receptor complex. transfer from CD14 to TLR4 and MD-2. J Biol Chem. 2001, 276: 21129-21135. 10.1074/jbc.M009164200.View ArticleGoogle Scholar
- Hayashi F, Smith KD, Ozinsky A, Hawn TR, Yi EC, Goodlett DR, Eng JK, Akira S, underhill DM, Aerem A: The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature. 2001, 410: 1099-1103. 10.1038/35074106.View ArticlePubMedGoogle Scholar
- Valdivia RH, Falkow S: Fluorescence-based isolation of bacterial genes expressed within host cells. Science. 1997, 277: 2007-2011. 10.1126/science.277.5334.2007.View ArticlePubMedGoogle Scholar
- DiDonato JA, Mercurio F, Karin M: Phosphorylation of I kappa B alpha precedes but is not sufficient for its dissociation from NF-kappa B. Mol Cell Biol. 1995, 15: 1302-1311.PubMed CentralView ArticlePubMedGoogle Scholar
- Norris FA, Wilson MP, Wallis TS, Galyov EE, Majerus PW: SopB, a protein required for virulence of Salmonella dublin, is an inositol phosphate phosphatase. Proc Natl Acad Sci U S A. 1998, 95: 14057-14059. 10.1073/pnas.95.24.14057.PubMed CentralView ArticlePubMedGoogle Scholar
- DiDonato JA: Assaying for I kappa B kinase activity. Methods Enzymol. 2000, 322: 393-400. 10.1016/S0076-6879(00)22038-7.View ArticlePubMedGoogle Scholar
- DiDonato JA, Hayakawa M, Rothwarf DM, Zandi E, Karin M: A cytokine-responsive IkappaB kinase that activates the transcription factor NF-kappaB. Nature. 1997, 388: 548-554. 10.1038/41493.View ArticlePubMedGoogle Scholar
- Hibi M, Lin A, Smeal T, Minden A, Karin M: Identification of an oncoprotein and UV-responsive protein kinase that binds and potentiates the c-Jun activation domain. Genes Dev. 1993, 7: 2135-2148.View ArticlePubMedGoogle Scholar
- Galkin VE, Orlova A, VanLoock MS, Zhou D, Galan JE, Egelman EH: The bacterial protein SipA polymerizes G-actin and mimics muscle nebulin. Nat Struct Biol. 2002, 9: 518-521.PubMedGoogle Scholar
- Wallis TS, Wood M, Watson P, Paulin S, Jones M, Galyov E: Sips, Sops, and SPIs but not stn influence Salmonella enteropathogenesis. Adv Exp Med Biol. 1999, 473: 275-280.View ArticlePubMedGoogle Scholar
- Muzio M, Ni J, Feng P, Dixit VM: IRAK (Pelle) family member IRAK-2 and MyD88 as proximal mediators of IL-1 signaling. Science. 1997, 278: 1612-1615. 10.1126/science.278.5343.1612.View ArticlePubMedGoogle Scholar
- Takeuchi O, Akira S: MyD88 as a bottle neck in Toll/IL-1 signaling. Curr Top Microbiol Immunol. 2002, 270: 155-167.PubMedGoogle Scholar
- Medzhitov R, Preston-Hurlburt P, Kopp E, Stadlen A, Chen C, Ghosh S, Janeway CA: MyD88 is an adaptor protein in the hToll/IL-1 receptor family signaling pathways. Mol Cell. 1998, 2: 253-258. 10.1016/S1097-2765(00)80136-7.View ArticlePubMedGoogle Scholar
- 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.PubMedGoogle Scholar
- Lien E, Sellati TJ, Yoshimura A, Flo TH, Rawadi G, Finberg RW, Carroll JD, Espevik T, Ingalls RR, Radolf JD, Golenbock DT: Toll-like receptor 2 functions as a pattern recognition receptor for diverse bacterial products. J Biol Chem. 1999, 274: 33419-33425. 10.1074/jbc.274.47.33419.View ArticlePubMedGoogle Scholar
- Lien E, Means TK, Heine H, Yoshimura A, Kusumoto S, Fukase K, Fenton , Oikawa M, Qureshi N, Monks B, Finberg RW, Ingalls RR, Golenbock DT: Toll-like receptor 4 imparts ligand-specific recognition of bacterial lipopolysaccharide. J Clin Invest. 2000, 105: 497-504.PubMed CentralView ArticlePubMedGoogle Scholar
- Alexopoulou L, Holt AC, Medzhitov R, Flavell RA: Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature. 2001, 413: 732-738. 10.1038/35099560.View ArticlePubMedGoogle Scholar
- Chuang TH, Ulevitch RJ: Cloning and characterization of a sub-family of human toll-like receptors: hTLR7, hTLR8 and hTLR9. Eur Cytokine Netw. 2000, 11: 372-378.PubMedGoogle Scholar
- Devary Y, Rosette C, DiDonato JA, Karin M: NF-kappa B activation by ultraviolet light not dependent on a nuclear signal. Science. 1993, 261: 1442-1445.View ArticlePubMedGoogle Scholar
- Heguy A, Baldari CT, Macchia G, Telford JL, Melli M: Amino acids conserved in interleukin-1 receptors (IL-1Rs) and the Drosophila toll protein are essential for IL-1R signal transduction. J Biol Chem. 1992, 267: 2605-2609.PubMedGoogle Scholar
- Croston GE, Cao Z, Goeddel DV: NF-kappa B activation by interleukin-1 (IL-1) requires an IL-1 receptor-associated protein kinase activity. J Biol Chem. 1995, 270: 16514-16517. 10.1074/jbc.270.28.16514.View ArticlePubMedGoogle Scholar
- Akashi S, Nagai Y, Ogata H, Oikawa M, Fukase K, Kusumoto S, Kawasaki K, Nishijima M, Hayashi S, Kimoto M, Miyake K: Human MD-2 confers on mouse Toll-like receptor 4 species-specific lipopolysaccharide recognition. Int Immunol. 2001, 13: 1595-1599. 10.1093/intimm/13.12.1595.View ArticlePubMedGoogle Scholar
- Schromm AB, Lien E, Henneke P, Chow JC, Yoshimura A, Heine H, Latz E, Monks BG, Schwartz JA, Miyake K, Golenbock DT: Molecular genetic analysis of an endotoxin nonresponder mutant cell line: a point mutation in a conserved region of MD-2 abolishes endotoxin-induced signaling. J Exp Med. 2001, 194: 79-88. 10.1084/jem.194.1.79.PubMed CentralView ArticlePubMedGoogle Scholar
- Lillehoj EP, Kim H, Chun EY, Kim KC: Pseudomonas aeruginosa Stimulates Phosphorylation of the Epithelial Membrane Glycoprotein Muc1 and Activates MAP Kinase. Am J Physiol Lung Cell Mol Physiol. 2004Google Scholar
- Lillehoj EP, Kim BT, Kim KC: Identification of Pseudomonas aeruginosa flagellin as an adhesin for Muc1 mucin. Am J Physiol Lung Cell Mol Physiol. 2002, 282: L751-L756.View ArticlePubMedGoogle Scholar
- Karin M, Takahashi T, Kapahi P, Delhase M, Chen Y, Makris C, Rothwarf D, Baud V, Natoli G, Guido F, Li N: Oxidative stress and gene expression: the AP-1 and NF-kappaB connections. Biofactors. 2001, 15: 87-89.View ArticlePubMedGoogle Scholar
- Jobin C, Sartor RB: The I kappa B/NF-kappa B system: a key determinant of mucosalinflammation and protection. Am J Physiol Cell Physiol. 2000, 278: C451-C462.PubMedGoogle Scholar
- Wood MW, Jones MA, Watson PR, Siber AM, McCormick BA, Hedges S, Rosquist R, Wallis TE, Galyov EE: The secreted effector protein of Salmonella dublin, SopA, is translocated into eukaryotic cells and influences the induction of enteritis. Cell Microbiol. 2000, 2: 293-303. 10.1046/j.1462-5822.2000.00054.x.View ArticlePubMedGoogle Scholar
- McCormick BA: The use of transepithelial models to examine host-pathogen interactions. Curr Opin Microbiol. 2003, 6: 77-81. 10.1016/S1369-5274(02)00003-6.View ArticlePubMedGoogle Scholar
- Sakaguchi T, Kohler H, Gu X, McCormick BA, Reinecker HC: Shigella flexneri regulates tight junction-associated proteins in human intestinal epithelial cells. Cell Microbiol. 2002, 4: 367-381. 10.1046/j.1462-5822.2002.00197.x.View ArticlePubMedGoogle Scholar
- Kaiser P, Rothwell L, Galyov EE, Barrow PA, Burnside J, Wigley P: Differential cytokine expression in avian cells in response to invasion by Salmonella typhimurium, Salmonella enteritidis and Salmonella gallinarum. Microbiology. 2000, 146 (Pt 12): 3217-3226.View ArticlePubMedGoogle Scholar
- Mizel SB, Snipes JA: Gram-negative flagellin-induced self-tolerance is associated with a block in interleukin-1 receptor-associated kinase release from toll-like receptor 5. J Biol Chem. 2002, 277: 22414-22420. 10.1074/jbc.M201762200.View ArticlePubMedGoogle Scholar
- Fitzgerald KA, Palsson-McDermott EM, Bowie AG, Jefferies CA, Mansell AS, Brady G, Brint E, Dunne A, Gray P, harte MT, McMurray D, Smith DE, Sims JE, Bird TA, O'Neill LA: Mal (MyD88-adapter-like) is required for Toll-like receptor-4 signal transduction. Nature. 2001, 413: 78-83. 10.1038/35092578.View ArticlePubMedGoogle Scholar
- Horng T, Barton GM, Medzhitov R: TIRAP: an adapter molecule in the Toll signaling pathway. Nat Immunol. 2001, 2: 835-841. 10.1038/ni0901-835.View ArticlePubMedGoogle Scholar
- Mizel SB, West AP, Hantgan RR: Identification of a sequence in human toll-like receptor 5 required for the binding of Gram-negative flagellin. J Biol Chem. 2003, 278: 23624-23629. 10.1074/jbc.M303481200.View ArticlePubMedGoogle Scholar
- Murthy KG, Deb A, Goonesekera S, Szabo C, Salzman AL: Identification of conserved domains in Salmonella muenchen flagellin that are essential for its ability to activate TLR5 and to induce an inflammatory response in vitro. J Biol Chem. 2003Google Scholar
- Smith KD, Andersen-Nissen E, Hayashi F, Strobe K, Bergman MA, Barrett SL, Cookson BT, Aderem A: Toll-like receptor 5 recognizes a conserved site on flagellin required for protofilament formation and bacterial motility. Nat Immunol. 2003, 4: 1247-1253. 10.1038/ni1011.View ArticlePubMedGoogle Scholar
- Yamaguchi S, Fujita H, Sugata K, Taira T, Iino T: Genetic analysis of H2, the structural gene for phase-2 flagellin in Salmonella. J Gen Microbiol. 1984, 130 (Pt 2): 255-265.PubMedGoogle Scholar
- Williams AW, Yamaguchi S, Togashi F, Aizawa SI, Kawagishi I, Macnab RM: Mutations in fliK and flhB affecting flagellar hook and filament assembly in Salmonella typhimurium. J Bacteriol. 1996, 178: 2960-2970.PubMed CentralPubMedGoogle Scholar
- Rock FL, Hardiman G, Timans JC, Kastelein RA, Bazan JF: A family of human receptors structurally related to Drosophila Toll. Proc Natl Acad Sci U S A. 1998, 95: 588-593. 10.1073/pnas.95.2.588.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.