The Cytosolic Pattern Recognition Receptor NOD1 Induces Inflammatory Interleukin-8 during Chlamydia trachomatis Infection

Abstract

Inflammation is a hallmark of chlamydial infections, but how inflammatory cytokines are induced is not well understood. Pattern recognition receptors (PRR) of the host innate immune system recognize pathogen molecules and activate intracellular signaling pathways that modulate immune responses. The role of PRR such as Toll-like receptors (TLR) and nucleotide-binding oligomerization domain (NOD) proteins in the endogenous interleukin-8 (IL-8) response induced during Chlamydia trachomatis infection is not known. We hypothesized that a PRR is essential for the IL-8 response induced by C. trachomatis infection. RNA interference was used to knock down the TLR signaling partner MyD88 as well as NOD1 and its signaling molecule receptor-interacting protein 2 (RIP2). IL-8 induced at 30 h postinfection by C. trachomatis was dependent on NOD1 signaling through RIP2; however, the IL-8 response was independent of MyD88-dependent TLR signaling. Activation of the extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase cellular signaling pathway, which is essential for up-regulation of IL-8 in response to C. trachomatis infection, was independent of NOD1 or RIP2. We conclude that the endogenous IL-8 response induced by C. trachomatis infection is dependent upon NOD1 PRR signaling through RIP2 as part of a signal system requiring multiple inputs for optimal IL-8 induction. Since ERK is not activated through this pathway, a concomitant interaction between the host and bacteria is additionally required for full activation of the endogenous IL-8 response.

Diseases cause by chronic or repeated Chlamydia infection are due to inflammation-mediated tissue damage (29, 30). Inflammatory factors are known to be up-regulated by epithelial cells during Chlamydia infection, although the exact mechanism by which this response is activated is not understood (11, 26). The most highly induced inflammatory cytokine during C. trachomatis infection is interleukin-8 (IL-8), an activator and chemoattractant of neutrophils associated with immunopathology (23, 28). IL-8 transcript levels are significantly up-regulated at 15 h postinfection (hpi) and are dependent on chlamydial protein synthesis (4). While extracellular IL-1α is involved in inducing IL-8 late in the developmental cycle at 72 hpi, prior to this time IL-8 production occurs only within inclusion-containing cells independent of trans-acting factors in the supernatant (4, 26). This has been termed the “endogenous” IL-8 response to differentiate it from the late, IL-1α-induced response.

The IL-8 response is controlled at the level of transcription through multiple promoter elements (14). The factors that bind to these elements are regulated by cellular signaling pathways (3, 15, 27, 35), of which NF-κB and extracellular signal-regulated kinase (ERK) mitogen-activated protein (MAP) kinase have been shown to be necessary for the endogenous IL-8 response to C. trachomatis infection (4, 5). Nuclearly localized intracellular IL-1α plays a role in IL-8 production as well, although the mechanism by which this occurs is as yet not known (6). While the signaling pathways which lead to IL-8 production are understood, the interactions that occur between Chlamydia and the host cell to induce activation of these pathways leading to inflammatory IL-8 are not characterized.

Inflammatory responses can be induced by activation of the innate immune system through identification of infection by germ line-encoded pattern recognition receptors (PRR) (19). These PRR induce host responses such as inflammation to activate and modulate the immune system (19). PRR include the transmembrane Toll-like receptors (TLR) and cytoplasmically localized nucleotide-binding oligomerization domain (NOD) proteins (2). Previous studies have found a significant role for NOD1 induction of inflammation during C. muridarum and C. pneumoniae infection, as well as with C. trachomatis infection of mouse cells (25, 34). In vivo, TLR2 has an important role in pathology caused by infection in mice (8, 9). While TLR and NOD proteins can activate the signaling pathways known to be essential for the endogenous IL-8 response to C. trachomatis infection, it is unknown whether a PRR is responsible for the ERK MAP kinase- and bacterial protein synthesis-dependent IL-8 induced in the context of an epithelial cell infection.

The purpose of this study was to determine the function of innate immune receptors in the inflammatory IL-8 response to Chlamydia infection of epithelial cells. The important role that PRR play in induction of the immune responses makes understanding their involvement in C. trachomatis infection essential to understand the host-bacterium interactions that are taking place. Using RNA interference (RNAi) to knock down PRR in HeLa 229 epithelial cells, we found that MyD88-dependent TLR signaling is not necessary for IL-8 induction after C. trachomatis infection. However, the cytosolic PRR NOD1 and its signaling partner receptor-interacting protein 2 (RIP2) were implicated in the endogenous IL-8 response. The ERK MAP kinase signaling pathway which is activated by C. trachomatis infection and is necessary for IL-8 up-regulation (5) was not inhibited by NOD1 or RIP2 knockdown, suggesting that their involvement in IL-8 production was not through the ERK pathway. Thus, we conclude that Chlamydia-induced IL-8 production is partially due to activation of the NOD1 receptor and its signaling partner RIP2; however, an independent signal for ERK is required for the full IL-8 response.

MATERIALS AND METHODS

Cell culture, growth, and isolation of Chlamydia.

HeLa 229 cervical epithelioid cells were grown in RPMI 1640 plus l-glutamine (Invitrogen, Carlsbad, CA) with 5% heat-inactivated fetal bovine serum (HyClone, Logan, UT) and 10 μM HEPES (Invitrogen) at 37°C in 5% CO₂ as described previously (4). Cell lines and bacterial stocks were tested for Mycoplasma contamination by PCR (VenorGeM Mycoplasma detection system; Sigma) prior to and throughout use and were found to be negative. C. trachomatis L2/434/Bu was cultured in L929 murine fibroblast cells and isolated as previously described (4). A dose of 2 inclusion-forming units of C. trachomatis L2/434/Bu per cell achieved greater than 90% infection of cells. Infections were performed for 2 h at 25°C in cell culture medium, cells were rinsed once with Hanks balanced salt solution, and medium was added to cells and incubated at 37°C in 5% CO₂.

RNAi.

Subconfluent HeLa cells in a six-well plate were transfected with small interfering RNA (siRNA) purchased from Ambion or a negative (nonspecific) siRNA (Dharmacon, Lafayette, CO) control using Oligofectamine (Invitrogen) (Silencer predesigned siRNAs 2680 CARD4 [NOD1], 455 RIPK2 [RIP2], and 242558 MYD88). Oligofectamine and Opti-MEM medium (Invitrogen) at a 1:8 ratio were vortexed for 1 s and incubated for 5 min at room temperature. Twenty-seven microliters of the Oligofectamine mix was added to 600 pmol/well siRNA in 200 μl Opti-MEM, vortexed for 10 s, and incubated for 20 min at room temperature, and then 230 μl of this RNA mixture was added to HeLa cells in six-well plates with 770 μl Opti-MEM. After 4 h, RPMI medium plus 30% fetal bovine serum (without antibiotics) was added. The transfection was repeated after 48 h. At 24 h after the second transfection, cells were split into 24-well plates for infection.

Transfection of TriDAP.

MurNAc-l-Ala-γ-d-Glu-meso-diaminopimelic acid (M-TriDAP) (InvivoGen, San Diego, CA) was transfected into cells as previously described (12). Five hundred microliters of RPMI 1640 (Invitrogen) with fetal bovine serum and HEPES, as described above, was added to HeLa cells in 24-well plates with or without 50 μg/ml M-TriDAP. One microgram of plasmid DNA in 50 μl Opti-MEM medium (Invitrogen) was mixed with 2 μl Lipofectamine (Invitrogen) in 50 μl Opti-MEM, incubated for 20 min, and added to cells in medium alone or medium with TriDAP.

QPCR.

Cell monolayers were rinsed with Hanks balanced salt solution, and RNA was extracted using Trizol reagent (Invitrogen). Total RNA was isolated according to the manufacturer's instructions and cDNA produced for quantitative PCR (QPCR) analysis as described by Buchholz and Stephens (4), using primers NOD1 F (5′ GCG AGG CGG GAC TAT CAG 3′), NOD1 R (5′ GCG GGT ACT TAG GGA GTT TGC 3′), RIP2 F (5′ CCA TGA AAA TAG TGG TTC TCC TGA A 3′), RIP2 R (5′ TCC AGG ACA GTG ATG CAG CTT 3′), MyD88 F (5′ CCA ACC TTC AGC AGT GAC AAG TC 3′), and MyD88 R (5′ AGC ATG ACC ACA GGC ATC CT 3′). QPCR to quantify chlamydial genomes was performed as described by Buchholz and Stephens (5) on the Smartcycler (Cepheid, Sunnyvale, CA).

Immunoblotting.

Infections were performed in six-well plates (Becton Dickinson, NJ). Cell lysate was harvested with sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis buffer, boiled, run on 10% SDS-polyacrylamide gels, and transferred to nitrocellulose for probing with antibodies specific to phosphorylated or total signaling pathway members. Phospho-p44/42 (ERK) MAP kinase E10 monoclonal antibody and p44/42 (ERK) MAP kinase antibody were purchased from Cell Signaling Technology (Danvers, MA).

Quantification of IL-8 in cell supernatants by ELISA.

IL-8 protein in cell culture supernatants was quantified as previously described (4). Briefly, plates were coated with monoclonal anti-human IL-8 (CXCL8) antibody (R&D Systems, Minneapolis, MN), at 0.5 μg/ml in phosphate-buffered saline (PBS). The blocking solution contained 5% (wt/vol) sucrose, 1% (wt/vol) bovine serum albumin (Fisher, Fair Lawn, NJ), and 0.05% (wt/vol) NaN₃ (sodium azide) in PBS containing 0.05% Tween 20. The detecting antibody, biotinylated polyclonal anti-human IL-8 (R&D Systems, Minneapolis, MN), was used at 20 ng/ml. Streptavidin-horseradish peroxidase (R&D Systems, Minneapolis, MN) was diluted in PBS, and 1,2-phenylenediamine dihydrochloride (DAKO Cytomation, Glostrup, Denmark) was used as the developing reagent. Standard curves were produced using recombinant human IL-8 (R&D Systems, Minneapolis, MN). The optical density of each well at 492 nm was then determined (Multiscan MCC/340 plate reader; Titertek, Huntsville, AL) and the pg/ml IL-8 determined by comparing the absorbance of the sample to a standard curve.

Statistical methods.

The statistical significance of data was determined using a two-tailed Student t test.

RESULTS

C. trachomatis-induced IL-8 is independent of MyD88-dependent TLR signaling.

Inflammation induced by pathogens is often through identification of infection by host PRR, including TLR (19). TLR activate cellular signaling pathways through TIR domain-mediated interactions with the adaptor molecule MyD88 (19). Almost all TLR signaling occurs through the conserved MyD88-dependent cascades (2); thus, the involvement of the TLR family of PRR in the host response to C. trachomatis infection can be examined by interrupting the function of the MyD88 adaptor molecule. To investigate the role of TLR in inducing the endogenous IL-8 response to C. trachomatis at 30 hpi, RNAi was used to knock down MyD88 levels in HeLa cervical epithelial cells. Protein levels of MyD88 are insufficient to be detected by immunoblotting, so changes in protein expression could not be measured. However, QPCR assessment of the MyD88 transcript showed a twofold reduction with siRNA treatment specific to MyD88 compared to that in cells treated with negative, nonspecific siRNA (data not shown).

To show functional inhibition of MyD88, the ability of IL-1 to induce IL-8 after siRNA treatment was determined. The IL-1 receptor activates signaling pathways and inflammatory cytokines such as IL-8 by interaction with MyD88 through TIR domains (24). After siRNA knockdown of MyD88, HeLa cells were treated with IL-1α, and IL-8 mRNA levels quantified by QPCR. IL-1α induction of IL-8 was inhibited 60% when MyD88 levels were reduced (Fig. 1), demonstrating the inhibited signaling activity and functionality of this adaptor molecule in the cells. Upon infection with C. trachomatis, however, there was no reduction in IL-8 mRNA levels measured at 30 hpi in cells treated with siRNA to MyD88 compared to those treated with a nonspecific siRNA (Fig. 1). Therefore, there is no role for MyD88-dependent TLR signaling in the endogenous IL-8 response induced by C. trachomatis 30 h after infection of epithelial cells.

FIG. 1.

IL-8 mRNA response after RNAi knockdown of MyD88. HeLa cells were treated with nonspecific (negative) siRNA or siRNA to MyD88 as described in Materials and Methods. Two hundred picograms of IL-1α was added 3 h prior to harvest. Total RNA was harvested at 30 hpi from triplicate wells of MyD88-specific siRNA- and nonspecific siRNA-treated uninfected, infected, and IL-1-stimulated cells. cDNA was produced using random primers and the relative change (n-fold) in IL-8 mRNA compared to that in uninfected and nonspecific siRNA-treated cells found by QPCR as described in Materials and Methods. Shown are the averages ± standard deviations. Data are from one of two experiments. *, P < 0.01.

NOD1 and RIP2 are necessary for the induction of inflammatory IL-8 by C. trachomatis.

While there is no role for MyD88-dependent TLR signaling in the induction of the endogenous IL-8 response to C. trachomatis infection, the host cell expresses other PRR that are involved in inducing inflammatory responses. NOD1 senses fragments of peptidoglycan in the cytosolic compartment to induce the innate immune response and activate cell signaling pathways which lead to IL-8 (12, 17). In other Chlamydia infection systems, NOD1 has been found to be important in stimulating inflammation (25, 34). To determine if NOD1 is involved in the induction of IL-8 during C. trachomatis infection, siRNA was used to knock down its levels within HeLa cells. Treatment with siRNA to NOD1 resulted in a two- to threefold drop in mRNA levels (data not shown). This reduction in NOD1 markedly decreased the level of the IL-8 transcript induced after C. trachomatis infection (Fig. 2A), indicating that NOD1 is necessary for the IL-8 response. To show functional inhibition of NOD1, the NOD1 ligand TriDAP was transfected into the cells and was found to induce IL-8 mRNA at 3 h posttransfection (Fig. 2B). This IL-8 response was reduced by RNAi treatment, similar to the profile seen after infection with C. trachomatis (Fig. 2A), showing that NOD1 function is inhibited by specific RNAi knockdown of NOD1 transcripts. The addition of IL-1α to cells resulted in the same levels of IL-8 mRNA, however (Fig. 2C), demonstrating no change in the ability of cells to produce IL-8 via other pathways after RNAi treatment.

FIG. 2.

IL-8 mRNA response after RNAi knockdown of NOD1 and RIP2. HeLa cells were treated with nonspecific (negative) siRNA or siRNA to NOD1 or RIP2 as described in Materials and Methods. Total RNA was harvested at 30 hpi from triplicate wells of each treatment. The relative change (n-fold) of IL-8 mRNA was found as described for Fig. 1. (A) IL-8 induced by C. trachomatis at 30 hpi. Shown are the averages ± standard deviations. Data are from one of three experiments. *, P < 0.01. (B) IL-8 mRNA induced 3 h after transfection with TriDAP as described in Materials and Methods. Shown are the averages ± standard deviations. Data are from one of three experiments. *, P < 0.05. (C) IL-8 mRNA induced 3 h after addition of 200 pg/ml IL-1α. Shown are the averages ± standard deviations. Data are from one of three experiments.

IL-8 is a transcriptionally controlled cytokine (14), and thus the IL-8 protein secreted should be reflective of the pattern of IL-8 mRNA induced. To confirm that RNAi knockdown of NOD1 also reduced the IL-8 protein secreted after C. trachomatis infection, IL-8 in supernatants of uninfected and infected cells with siRNA treatments was quantified. Consistent with the IL-8 mRNA profile, less IL-8 protein was secreted by infected cells with NOD1 levels reduced than by infected cells with negative (nonspecific) siRNA treatment (Fig. 3A). Thus, the cytosolic innate immune receptor NOD1 is necessary for optimal induction of IL-8 by C. trachomatis at 30 hpi.

FIG. 3.

IL-8 protein secretion after RNAi knockdown of NOD1 and RIP2. (A) HeLa cells were treated with nonspecific siRNA or siRNA to NOD1 or RIP2 as described in Materials and Methods. Supernatant from triplicate wells was analyzed by ELISA as described in Materials and Methods. Shown are the averages ± standard deviations. *, P < 0.01. (B) Measurement of Chlamydia infection in siRNA-treated cells at 30 hpi by QPCR as described in Materials and Methods. Shown are the averages ± standard deviations for three separate wells. Data are from one of two experiments.

NOD1 and NOD2 act to induce inflammation by stimulation of cellular signaling pathways through the adaptor molecule RIP2 (RICK, CARDIAK), which induces NF-κB as well as Jun N-terminal protein kinase and ERK MAP kinase (21, 22, 32). To determine if the IL-8 up-regulated during C. trachomatis infection is induced through the signaling adaptor molecule RIP2, RNAi was used to reduce RIP2 levels in HeLa cells prior to infection. Treatment with siRNA to RIP2 resulted in a two- to threefold drop in mRNA levels (data not shown). As with the diminished IL-8 response when NOD1 was reduced by RNAi, knockdown of the adaptor molecule RIP2 resulted in less IL-8 mRNA in response to infection and TriDAP treatment (Fig. 2A and B), whereas there was no difference in the response to IL-1α induction of IL-8 (Fig. 2C). RIP2 involvement in stimulation of IL-8 after infection was also shown by reduced IL-8 protein secretion after C. trachomatis infection (Fig. 3A). We conclude that the signaling partner of NOD1, RIP2, is necessary for the induction of IL-8 in response to C. trachomatis infection.

IL-8 is induced in a growth-dependent manner (4), and thus any reduction in or interference with C. trachomatis growth could reduce the amount of IL-8 produced. To determine if RNAi treatments are affecting infection or growth by Chlamydia, chlamydial genomic copies were measured by quantitative PCR analysis as previously described (5). No difference in Chlamydia genome copies was found between infected cells with NOD1, RIP2, and MyD88 siRNA treatments and those with negative nonspecific siRNA treatment (Fig. 3B). Altogether, we conclude from these data that NOD1 activation and signaling through RIP2 during C. trachomatis infection of epithelial cells is, in part, responsible for induction of the inflammatory IL-8 response.

NOD1 and RIP2 activation of the IL-8 response to C. trachomatis infection is independent of the ERK MAP kinase pathway.

NOD1 induces inflammation and cytokines such as IL-8 by activation of host cell signaling pathways such as ERK MAP kinase through RIP2 (13, 18, 21). ERK is activated during C. trachomatis infection concurrent with IL-8 up-regulation and is essential for the production of IL-8 (5). To determine whether NOD1 activates the ERK pathway during infection, C. trachomatis-infected cells with NOD1 or RIP2 reduced by RNAi treatment were analyzed for ERK phosphorylation. Antibodies specific to the phosphorylated ERK1/2 proteins or total ERK1/2 proteins were used to determine the activation of the pathway at 30 h after C. trachomatis infection. RNAi knockdown of NOD1 and RIP2 reduced the IL-8 response at 30 h after C. trachomatis infection (Fig. 2 and 3). However, while the ERK MAP kinase pathway was activated at 30 hpi (Fig. 4), reduced levels of NOD1 and RIP2 did not diminish this phosphorylation of ERK (Fig. 4). We conclude that C. trachomatis activation of ERK is not occurring through NOD1 and RIP2 induction of the pathway. These data demonstrate that while NOD1 is necessary for the optimal IL-8 response during C. trachomatis infection, the role of NOD1 in inducing IL-8 is parallel to activation of the ERK pathway and not in series through activation of the ERK pathway.

FIG. 4.

Phosphorylation of ERK1/2 after C. trachomatis infection of cells with RNAi knockdown of NOD1 and RIP2. HeLa cells were treated with nonspecific (negative) siRNA or siRNA to NOD1 or RIP2 as described in Materials and Methods. Total protein was harvested at 30 hpi, run on an SDS-polyacrylamide gel, and probed with antibodies specific for total and phosphorylated ERK as described in Materials and Methods. Data are from one of two separate experiments.

DISCUSSION

Bacterial pathogen-associated molecules such as flagellin, lipopolysaccharide (LPS), and CpG DNA are recognized by the innate immune system through PRR such as TLR and NOD proteins to induce inflammatory responses, including the cytokine IL-8 (1, 2, 19). IL-8 is highly induced during C. trachomatis infection of epithelial cells (4, 26); however, the function of TLR and NOD proteins in mediating the endogenous IL-8 response is not known. This investigation sought to understand what role TLR and NOD PRR play in the cellular IL-8 response to Chlamydia infection in epithelial cells.

IL-8 induction at 30 hpi, prior to the late 72-hpi activity induced by extracellular IL-1α (4), was dependent on NOD1 and its signaling partner RIP2 but independent of MyD88 TLR function. Human NOD1 is the cytosolic receptor for the peptidoglycan fragments GlcNAc-TriDAP, which are produced by gram-negative bacteria (12). Sensing of these peptidoglycan fragments in the cytosol of a cell by NOD1 indicates a bacterial infection and precipitates induction of an inflammatory immune response through RIP2 activation of signaling pathways (18). NOD1 stimulates NF-κB but also can activate the Jun N-terminal protein kinase and ERK MAP kinase pathways (13, 20, 21). ERK MAP kinase is known to be activated by C. trachomatis infection and is necessary for the growth-dependent induction of IL-8 (5). While NOD1 and RIP2 were necessary for inducing the endogenous IL-8 response, they were not alone responsible for the ERK phosphorylation activated by C. trachomatis.

IL-8 up-regulation during Chlamydia infection of epithelial cells was shown in this study to be independent of MyD88. It is important to note that this was at 30 hpi, prior to the time at which extracellular IL-1α induces inflammation through the MyD88-dependent signaling of the IL-1 receptor (4, 26). It was recently demonstrated during C. trachomatis infection that intracellular IL-1α is necessary for IL-8 induction independently of signaling through the IL-1 receptor (6). The mechanism is not known, but it is suggested that intracellular IL-1α may be acting via modulation of histones (6), which could act in conjunction with transcription factors induced by cellular signaling pathways (4, 5). Stimulation by IL-1α through the IL-1 receptor is significant in the manifestation of disease, as shown in an ex vivo model of fallopian tube tissue destruction with C. trachomatis infection (16). However, our focus in this study was the initial interactions between the host cell and bacteria, and we find that these differ from the MyD88-dependent processes that are significant later in the course of infection. Thus, while later amplification of the inflammatory processes is dependent on MyD88 function, this work demonstrates that it is not involved in early and initial induction of the IL-8 response.

The function of NOD1 has been investigated in mouse infections with C. muridarum and C. trachomatis (34) and in endothelial infections with C. pneumoniae (25). Consistent with our results that NOD1 induces IL-8 in human HeLa epithelial cells, NOD1 was shown to also be involved in inducing the inflammatory cytokines IL-6 and macrophage inflammatory protein 2 in mouse cells (34) and IL-8 in human endothelial cells infected with C. pneumoniae. The role found for NOD1 in activation of inflammatory cytokines in response to C. trachomatis infection has interesting implications beyond sensing intracellular pathogens. Previously, peptidoglycan was unable to be isolated from Chlamydia elementary bodies (7). Sequencing of the genome found that Chlamydia does possess genes for peptidoglycan synthesis (31), and thus it is logical to presume that Chlamydia produces peptidoglycan at some time during its developmental cycle. Indirectly, this study adds to the evidence that Chlamydia produces peptidoglycan as NOD1 is activated during infection (25, 34). The mechanism by which NOD1 gains access to peptidoglycan fragments is unknown. Chlamydia spends its entire developmental cycle within the inclusion, separated from the host cell cytosol. How fragments of the bacterial cell wall might make their way out of the inclusion is unknown. The activation of NOD1 during Chlamydia infection indicates greater exchange between chlamydial organisms or the lumen of the inclusion with the host cell cytosol than previously appreciated.

MyD88-dependent TLR interactions were characterized in this study, and we can rule out almost all TLR as activating IL-8 in response to C. trachomatis infection of epithelial cells. The TLR that does not signal via MyD88, TLR3, is unlikely to be involved in sensing a Chlamydia infection, as TLR3 senses double-stranded RNA in the context of a viral infection (20). Alternatives to MyD88-dependent signaling have been characterized (2, 20). TRIF associates with TLR4 or TLR3 to induce MyD88-independent signaling; however, TLR4 is not functional in the HeLa 229 cells used in this study (20). While we have not assessed the function of every host PRR, we have characterized two major classes of PRR and their function in IL-8 induction by C. trachomatis.

TLR2 has been identified as essential for oviduct pathology in mouse infection models and for maximal expression of inflammatory cytokines (8, 9), indicating a role for TLR2 in inflammation induced by C. trachomatis infection. Epithelial cells are often refractory to extracellular inflammatory stimuli such as LPS activation of TLR4, presumably because of their residence in locations of the body with natural microbial flora. HeLa 229 cervical epithelial cells do not express TLR2 or respond to LPS stimuli (reference 10 and data not shown), consistent with this refractory characteristic. Thus, in this in vitro system, TLR2 cannot be responsible for the inflammation induced, consistent with data showing no role for MyD88 in the IL-8 response.

During in vivo C. trachomatis infection, both TLR2-dependent and -independent mechanisms of IL-8 induction are likely to be elicited. While the cells used in this system do not express TLR2, studies have found that TLR2 and NOD1 can act synergistically to induce inflammation (33). Dual recognition through TLR and NOD proteins may enhance responses to infection and be important in activation of the innate immune response. However, it should be noted that NOD1 knockout mice have no change in the duration of Chlamydia infection or level of cytokine induction (34). The overlapping functions of, and not synergistic activation through, TLR2 and NOD1 may explain the apparent undetectable role for NOD1 in mouse systems in vivo. Characterization of the in vitro IL-8 induction defines the host-pathogen interactions upon infection of the host cell that lead to inflammation as well the mechanisms by which Chlamydia influences host cell signaling and immune responses.