Nonpathogenic enteric bacterial species initiate and perpetuate experimental colitis in IL-10 gene-deficient mice (IL-10−/−). Bacteria-specific effects on the epithelium are difficult to dissect due to the complex nature of the gut microflora. We showed that IL-10−/− mice compared with wild-type mice fail to inhibit proinflammatory gene expression in native intestinal epithelial cells (IEC) after the colonization with colitogenic Gram-positive Enterococcus faecalis. Interestingly, proinflammatory gene expression was transient after 1 wk of E. faecalis monoassociation in IEC from wild-type mice, but persisted after 14 wk of bacterial colonization in IL-10−/− mice. Accordingly, wild-type IEC expressed phosphorylated NF-κB subunit RelA (p65) and phosphorylated Smad2 only at day 7 after bacterial colonization, whereas E. faecalis-monoassociated IL-10−/− mice triggered persistent RelA, but no Smad2 phosphorylation in IEC at days 3, 7, 14, and 28. Consistent with the induction of TLR2-mediated RelA phosphorylation and proinflammatory gene expression in E. faecalis-stimulated cell lines, TLR2 protein expression was absent after day 7 from E. faecalis-monoassociated wild-type mice, but persisted in IL-10−/− IEC. Of note, TGF-β1-activated Smad signaling was associated with the loss of TLR2 protein expression and the inhibition of NF-κB-dependent gene expression in IEC lines. In conclusion, E. faecalis-monoassociated IL-10−/−, but not wild-type mice lack protective TGF-β/Smad signaling and fail to inhibit TLR2-mediated proinflammatory gene expression in the intestinal epithelium, suggesting a critical role for IL-10 and TGF-β in maintaining normal epithelial cell homeostasis in the interplay with commensal enteric bacteria.

The normal mucosal immune system acquires tolerance (hyporesponsiveness) to enteric indigenous flora, while protective cell-mediated and humoral immune responses to enteropathogens are maintained. This complex homeostasis presents an intriguing immunological paradox, and is broken under conditions of chronic intestinal inflammation, including ulcerative colitis and Crohn’s disease (1, 2). Microbial agents of the normal enteric microflora are involved in each of the current etiologic theories of these chronically relapsing, immunologically mediated idiopathic disorders (3, 4). Duchmann et al. (5, 6, 7) reported a loss of immunologic tolerance in active inflammatory bowel diseases (IBD)4 with mucosal T lymphocytes proliferating in response to commensal enteric bacteria. These findings are consistent with the clinical observation that inflammation in bypassed distal ileal or colonic segments of Crohn’s disease patients is absent after proximal diversion of the fecal stream (8, 9), but immune responsiveness and inflammation are reactivated within 1 wk of perfusing ileostomy effluent into the bypassed ileum (10). In addition, the selective impact of certain nonpathogenic commensal luminal enteric bacteria on the development of chronic intestinal inflammation is elegantly illustrated in comparative studies using germfree and gnotobiotic rodent models for experimental colitis (11). For example, reconstitution studies with various nonpathogenic bacteria implicate Enterococcus faecalis as particularly important to the induction of colitis in selectively colonized gnotobiotic IL-10 gene-deficient (IL-10−/−) mice. IL-10−/− 129 SvEv mice developed experimental colitis after 12–16 wk of E. faecalis monoassociation (12, 13). The absence of colitis and pathologic immune responses in E. faecalis-colonized wild-type (WT) mice demonstrated the nonpathogenic nature of this Gram-positive enteric bacterial species, and, most importantly, suggests that normal hosts develop immunosuppressive mechanisms that control mucosal immune responses to the constant challenge of commensal bacterial Ags.

Intestinal epithelial cells (IEC), which isolate the host from the gut luminal environment, constitutively express or can be induced to express TLR, costimulatory molecules, components of the human MHC, and a wide range of inflammatory and chemoattractant cytokines when activated by enteric pathogens or inflammatory products (14, 15, 16, 17). Most of these molecules are in part transcriptionally regulated by the transcription factor NF-κB (18). We demonstrated that nonpathogenic Gram-negative Bacteroides vulgatus induce RelA (NF-κB p65 subunit) phosphorylation and NF-κB activation in vivo in the epithelium as well as in primary and IEC lines (19, 20). Monoassociation of germfree Fisher rats with B. vulgatus induced transient nuclear localization of phosphorylated RelA in the intestinal epithelium in a TLR4- and NF-κB-dependent manner, demonstrating the physiological relevance of the NF-κB signaling cascade in Gram-negative bacterial communication with IEC. In separate studies, Hornef et al. (21) showed that Escherichia coli-derived LPS was internalized by murine epithelial cells to stimulate the IκB/NF-κB system via intracellular TLR4.

Although bacteria trigger host responses by multiple mechanisms, the cornerstone of innate signaling is mediated by a set of well-conserved pattern recognition receptors, including TLR, of which >10 isoforms have been identified (22, 23, 24). Specific microbial ligands have been assigned to at least five different molecules. TLRs are transmembrane proteins characterized by an extracellular domain containing leucine-rich repeats and an intracellular domain homologous to the IL-1R, or Toll/IL-1R (TIR). Ligand-specific binding to TLR promotes interaction of the cytoplasmic TIR domain with adaptor proteins, including the MyD88, followed by the recruitment of multiple kinases and activation of downstream target effector systems, including the MAPK as well as the IκB/NF-κB transcriptional system (24, 25). Of note, under conditions of chronic intestinal inflammation, increased expression of TLRs was shown in lamina propria macrophages (26) and epithelial cells (27). TLR2, either alone or in conjunction with TLR1 or TLR6, recognizes pathogen-associated molecular patterns from Gram-positive bacteria, including lipoteichoic acid, peptidoglycans, neisserial porins, bacterial tripalmytoylated, and mycoplasmal diacetylated lipoproteins as well as yeast products (24).

In this study, we characterized the molecular mechanisms of Gram-positive E. faecalis-induced signal transduction in native IEC from gnotobiotic WT and IL-10−/− mice. Most importantly, and in contrast to wild-type animals, we show that E. faecalis-monoassociated IL-10−/− mice lack protective TGF-β/Smad2 signaling and exhibit persistent RelA phosphorylation through the TLR2 cascade at early stages of bacterial colonization (days 3, 7, 14, and 28), well before the onset of histologic colitis. Proinflammatory gene expression was transiently induced after 1 wk of E. faecalis monoassociation in WT mice, but persisted after 14 wk of bacterial colonization in IL-10−/− IEC, suggesting that initial bacteria-epithelial cell interactions may contribute to the induction of immunopathology at the late/chronic phase of intestinal inflammation in the genetically susceptible host. Interestingly, TLR2 protein expression was absent after 7 days in WT mice, but persisted in IL-10−/− IEC. Important for the mechanistic basis of altered TLR/NF-κB signaling in IEC from WT vs IL-10−/− mice, we could demonstrate that TGF-β1-activated Smad signaling was directly associated with the loss of TLR2 protein expression and the inhibited E. faecalis-induced NF-κB activity as well as proinflammatory gene expression in Mode-K cells.

Germfree 129 SvEv TAC mice and germfree IL-10 gene-deficient (−/−) 129 SvEv TAC mice (derived by E. Balish, University of Wisconsin, Madison, WI) were monoassociated at 12–16 wk of age with a human oral isolate of E. faecalis (a generous gift from M. Huycki, University of Oklahoma, Oklahoma City, OK) and maintained in the Gnotobiotic Animal Core at the College of Veterinary Medicine, North Carolina State University. Bacterial monoassociation and absence of contamination by other bacterial species were confirmed by culturing samples from the small and large intestine at necropsy and culturing serial fecal samples. Animal use protocols were approved by the Institutional Animal Care and Use Committee, North Carolina State University. Mice were killed 3, 7, 14, and 28 days and 14 wk after initial bacterial colonization. Germfree mice were used as controls. Sections of the ileum, cecum, proximal, and distal colon were fixed in 10% neutral buffered Formalin. The fixed tissue was embedded in paraffin. Histology scoring (0–4) was analyzed by blindly assessing the degree of lamina propria mononuclear cell infiltration, crypt hyperplasia, goblet cell depletion, and architectural distortion, as previously described (28).

E. faecalis-monoassociated and germfree mice were euthanized, and the cecum as well as colon were removed and placed in DMEM (Invitrogen Life Technologies) containing 5% FCS. Cecum and colon were cut longitudinally, washed three times in calcium/magnesium-free HBSS (Invitrogen Life Technologies), cut into pieces 0.5 cm long, and incubated at 37°C in 40 ml of DMEM containing 5% FCS and 1 mM DTT for 30 min in an orbital shaker. The supernatant was filtered and centrifuged for 5 min at 400 × g, and the cell pellet was resuspended in DMEM containing 5% FCS. The remaining tissue was incubated in 30 ml of PBS (1×) containing 1.5 mM EDTA for additional 10 min. The supernatant was filtered and centrifuged for 5 min at 400 × g, and the cell pellet was resuspended in DMEM containing 5% FCS. Finally, primary IEC were collected by centrifugation through a 25/40% discontinuous Percoll gradient at 600 × g for 30 min. Cell viability and purity were assessed by trypan blue exclusion and FACS analysis using mouse anti-CD3 mAb (BD Pharmingen; clone G4.18). Cells were >80% viable and >90% pure. Primary mouse IEC from cecum and colon were combined and collected in sample buffer for subsequent RNA isolation as well as Western blot analysis.

The mouse IEC lines CMT-93 (passage 10–30) (ATCC CRL 223; American Type Culture Collection) and Mode-K (passage 10–30) (a generous gift from I. Autenrieth, University of Tübingen, Tübingen, Germany) were grown in a humidified 5% CO2 atmosphere at 37°C to confluency in six-well tissue culture plates (Cell Star; Greiner Bioscience), as previously described (20). We also transfected Mode-K cells with the following plasmids, according to the manufacturer’s instructions (InvivoGen): 1) pDUO vector (InvivoGen) for murine TLR4/MD-2 coexpressing TLR4 and the accessory molecule MD-2; 2) pZERO vector (InvivoGen) for murine TLR2ΔTIR lacking the cytoplasmic TIR domain. Stable transfected cells were then selected for their blasticidin (1 μg/ml) and puromycin (1 μg/ml) resistence, respectively. WT and TLR2−/− myoembryogenic fibroblast (MEF) were a generous gift from C. Kirschning (Technical University of Munich, Munich, Germany). E. faecalis was aerobically grown at 37°C in Luria broth containing tryptone (1%), yeast extract (0.5%), and NaCl (0.5%). Bacteria were harvested by centrifugation (3000 × g, 15 min) at stationary growth phase, washed in PBS (1× PBS, pH 7.4), and diluted in DMEM (Invitrogen Life Technologies). Confluent epithelial cell monolayers were infected with E. faecalis at a bacterium-to-epithelial cell ratio (multiplicity of infection (moi)) of 100 for various time points. Where indicated, Mode-K cells were treated with TGF-β1 (20 ng/ml; R&D Systems).

RNA from isolated native IEC and IEC lines was extracted using TRIzol reagent (Invitrogen Life Technologies), according to the manufacturer’s instructions. Extracted RNA was dissolved in 20 μl of water containing 0.1% diethyl-pyrocarbonate. For reverse transcription, 1 μg of total RNA was added to 30 μl of reaction buffer containing 8 μl of 5× first-strand buffer, 4 μl of DTT (100 mM), and 6 μl of desoxyribonucleoside triphosphate mixture (300 μM) (all reagents from Invitrogen Life Technologies), and incubated for 5 min at 65°C. After adding 10 μl of a solution containing 0.2 μg of random hexamers, 40 U of RNase Out, and 200 U of Moloney murine leukemia virus reverse transcriptase (all reagents from Invitrogen Life Technologies), the total mixture was incubated for an additional 60 min at 37°C, followed by a final 1-min heating step at 99°C.

Real-time PCR was performed in glass capillaries using a Light Cycler system (Roche Diagnostics). Primer sequences and amplicon sizes are provided in Table I. For real-time PCR, 1 μl of reverse-transcribed cDNA was added in a total volume of 10 μl of PCR buffer containing 1× LC-FastStart DNA Master Mix (Roche), MgCl2 (4 μM), and forward and reverse primers (20 μM). The PCR program was one cycle of denaturation at 95°C for 10 min, followed by 50 cycles of 95°C for 15 s, annealing at 60°C for 10 s, and extension at 72°C for 20 s. The amplified product was detected by the presence of a SYBR green fluorescent signal. Melting curve analysis and gel electrophoresis were used to document the amplicon specificity. Calibration curves were generated by measuring serial dilutions of stock cDNA to calculate the amplification efficiency (E). The crossing point (Cp) of the log-linear portion of the amplification curve was determined. The relative induction of gene mRNA expression was calculated using the following equation: EΔCp (control samples − treated samples) and normalized for the expression of GAPDH mRNA (29). Triplicate samples were measured in duplicates and blotted as fold increase between treated and untreated control samples.

Table I.

Primer sequences for real-time PCR

Forward PrimerReverse PrimerProduct Length (bp)
IL-6 5′-acaacgatgatgcactt-3′ 5′-cttggtccttagccact-3′ 334 
IP-10 5′-tccctctcgcaaggac-3′ 5′-ttggctaaacgctttcat-3′ 209 
TGF-β1 5′-cgccatctatgagaaaacc-3′ 5′-gtaacgccaggaattgt-3′ 190 
TGF-βR2 5′-catttggttccaaggtgc-3′ 5′-tggtagtgttcagcgag-3′ 282 
GAPDH 5′-atcccagagctgaacg-3′ 5′-gaagtcgcaggagaca-3′ 198 
Forward PrimerReverse PrimerProduct Length (bp)
IL-6 5′-acaacgatgatgcactt-3′ 5′-cttggtccttagccact-3′ 334 
IP-10 5′-tccctctcgcaaggac-3′ 5′-ttggctaaacgctttcat-3′ 209 
TGF-β1 5′-cgccatctatgagaaaacc-3′ 5′-gtaacgccaggaattgt-3′ 190 
TGF-βR2 5′-catttggttccaaggtgc-3′ 5′-tggtagtgttcagcgag-3′ 282 
GAPDH 5′-atcccagagctgaacg-3′ 5′-gaagtcgcaggagaca-3′ 198 

Purified primary IEC or E. faecalis-stimulated Mode-K cells were lysed in 1× Laemmli buffer, and 20–50 μg of protein was subjected to electrophoresis on 10% SDS-PAGE gels. Where indicated, IEC cells were pretreated for 1 h with 20 μM proteasome inhibitor MG132 (BIOMOL). Anti-IκBα (C21; Santa Cruz Biotechnology); anti-phospho-IκBα (Ser32), anti-phospho-RelA (Ser536), RelA, and anti-phospho-Smad2 (Ser465/467) (all Abs from Cell Signaling Technology); TLR2 (a generous gift from C. Kirschning); Smad1/2/3 (Santa Cruz Biotechnology); and anti-β-actin (Valeant Pharmaceuticals) were used to detect immunoreactive phospho-IκBα, total IκBα, phospho-RelA, RelA, phospho-Smad2, total Smad, TLR2, and β-actin, respectively, using an ECL light-detecting kit (Amersham), as previously described (20).

Mode-K cells were stimulated with E. faecalis for 30 min in the presence or absence of TGF-β1 (20 ng/ml). Mode-K cells were fixed by adding formaldehyde to a final concentration of 1%, and nuclear extraction and ChIP were performed by using a kit from Active Motif. As previously described (19), the cells were lysed after formaldehyde fixation, and chromatin of isolated nuclei was sheared by sonication. Extracts were normalized according to their DNA concentration. Immunoprecipitations were conducted overnight at 4°C using 5 μl of anti-phospho NF-κB p65 Ab (Cell Signaling Technology). Immune complexes were collected with salmon sperm-saturated protein A/G agarose for 30 min and washed three times in high salt buffer, followed by three washes with no salt buffer. DNA cross-links of the immune complexes were reverted by heating. After proteinase K digestion, the DNA was extracted with phenol-chloroform and precipitated in ethanol. DNA isolated from an aliquot of the total nuclear extract was used as a loading control for the PCR (input control). PCR was performed with total DNA (1 μl, input control) and immunoprecipitated DNA (1 μl) using the following IL-6 promoter-specific primers: IL-6A (5′) 5-GACATGCTCAAGTGCTGAGTCAC-3; IL-6B (3′) 5-AGATTGCACAATGTGACGTCG-3. The length of the amplified product was 125 bp. The PCR products (10 μl) were subjected to electrophoresis on 2% agarose gels.

IEC lines were infected overnight with adenoviral dominant-negative (dn) I-κB kinase (IKK)β (Ad5dnIKKβ) and Ad5IκBαAA vectors (a generous gift from C. Jobin, University of North Carolina, Chapel Hill, NC) in serum-reduced (2%) cell culture medium in the absence of antibiotics at an moi of 50, as previously described (20). Ad5Smad7 (a generous gift from D. Brenner, Columbia University, New York, NY) was used to infect Mode-K cells for 24 h. Expression of hemagglutinin (HA)- and FLAG-tagged mutant molecules was controlled by Western blot analysis using mouse anti-HA (Roche Molecular Biochemicals) and mouse anti-FLAG2 (Sigma-Aldrich) mAbs. Ad5GFP was used as a viral negative control. The adenovirus was removed by washing, and fresh cell culture medium was added. Cells were stimulated with E. faecalis (100 moi) for various times.

Protein concentrations were determined in spent culture supernatants of IEC cultures using an ELISA technique. IL-6 and IFN-γ-inducible protein-10 (IP-10) production were determined by mouse-specific ELISA kits, according to the manufacturer’s instructions (R&D Systems).

Data are expressed as the mean ± SD of triplicates. Statistical analysis was performed by the two-tailed Student’s t test for paired data and considered significant if p values were <0.05 (*) or <0.01 (**).

We investigated E. faecalis-induced proinflammatory gene expression and signal transduction in the intestinal epithelium of WT and IL-10−/− mice sequentially after bacterial colonization. First, E. faecalis-monoassociated WT and IL-10−/− mice (n = 3–5) were killed after 1 and 14 wk of initial bacterial colonization, and IP-10 mRNA expression was measured using Light Cycler real-time PCR. Luminal contents were plated, confirming equal colonization with E. faecalis between WT and IL-10−/− mice. Histological analysis was performed on paraffin-embedded tissue sections from the cecum and distal colon and RNA was isolated from native IEC of WT and IL-10−/− mice (cecum + colon). As shown in Fig. 1,A, real-time PCR analysis showed that E. faecalis monoassociation triggered IP-10 gene expression in both WT (7.3-fold increase) and IL-10−/− IEC (11.1-fold) after the first week of bacterial colonization, well before the onset of histologic colitis. Interestingly and consistent with the lack of histopathology (Fig. 1 B), IP-10 gene expression returned to low levels (2.5-fold) after 14 wk of E. faecalis monoassociation in WT mice. In contrast to WT mice, the initial induction of IP-10 gene expression in IL-10−/− IEC after 1 wk of bacterial colonization (11.1-fold) further increased after 14 wk of E. faecalis monoassociation (18.9-fold). E. faecalis-monoassociated IL-10−/− mice developed severe colitis after 14 wk of bacterial colonization (histological score of 3.8). IL-6 gene expression was not induced in IEC from WT nor IL-10−/− mice. Bacterial colonization of the cecum and colon was similar in WT and IL-10−/− mice (4–6 × 109 CFU/g luminal content).

FIGURE 1.

A and B, Differential IP-10 gene expression in native IEC from wild-type vs IL-10−/− mice. Germfree WT and IL-10−/− mice were colonized with E. faecalis at 12–16 wk of age. Mice (n = 3–5) were killed at days 3, 7, 14, and 28 as well as 14 wk after initial bacterial colonization, and native IEC from large intestine (cecum + colon) were isolated, as described in Materials and Methods. IEC from germfree mice were used as controls. A, Total RNA was extracted and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine IP-10 and GAPDH. The induction of IP-10 mRNA was calculated relative to germfree WT controls (mean fold increase ± SD) using the crossing point of the log-linear portion of the amplification curve after normalization with GAPDH. B, Blinded histological analysis (score 0–4) was performed in paraffin-embedded tissue sections of the cecum and distal colon, as described in Materials and Methods.

FIGURE 1.

A and B, Differential IP-10 gene expression in native IEC from wild-type vs IL-10−/− mice. Germfree WT and IL-10−/− mice were colonized with E. faecalis at 12–16 wk of age. Mice (n = 3–5) were killed at days 3, 7, 14, and 28 as well as 14 wk after initial bacterial colonization, and native IEC from large intestine (cecum + colon) were isolated, as described in Materials and Methods. IEC from germfree mice were used as controls. A, Total RNA was extracted and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine IP-10 and GAPDH. The induction of IP-10 mRNA was calculated relative to germfree WT controls (mean fold increase ± SD) using the crossing point of the log-linear portion of the amplification curve after normalization with GAPDH. B, Blinded histological analysis (score 0–4) was performed in paraffin-embedded tissue sections of the cecum and distal colon, as described in Materials and Methods.

Close modal

It is well established in various cell systems that Gram-positive bacteria trigger NF-κB activation and proinflammatory gene expression through the TLR2 signaling cascade (24). We next measured phospho-RelA, RelA, TLR2, and β-actin in isolated native IEC from WT and IL-10−/− mice after 3, 7, 14, and 28 days of monoassociation using Western blot analysis. As shown in Fig. 2 (upper blot), E. faecalis-monoassociated WT mice displayed transient RelA (p65) serine phosphorylation in native large IEC at days 3 (weak expression) and 7, but not at days 3, 14, and 28, or in germfree controls. Conversely, persistent RelA phosphorylation at days 3, 7, 14, and 28 was found in IEC of E. faecalis-monoassociated IL-10−/− mice. Interestingly, phospho-RelA levels steadily increased in IEC of E. faecalis-monoassociated IL-10−/− mice with maximal induction by 14 days after initial bacterial colonization. In contrast to WT mice, increased levels of phospho-RelA were detectable in IEC of germfree IL-10−/− mice. Of note, total RelA was equally expressed in IEC from WT and IL-10−/− mice. As shown in Fig. 2 (lower blot), TLR2 protein was present in IEC from germfree WT and IL-10−/− mice. Interestingly, while TLR2 protein expression was completely diminished in IEC from WT mice at days 14 and 28 of E. faecalis monoassociation, TLR2 expression persisted in IL-10−/− IEC. β-actin protein bands confirmed equal loading of samples (50 μg of protein).

FIGURE 2.

Differential RelA phosphorylation and TLR2 protein expression in native IEC from wild-type vs IL-10−/− mice. Germfree WT and IL-10−/− mice were colonized with E. faecalis at 12–16 wk of age. Mice (n = 3–5) were killed at days 3, 7, 14, and 28 as well as 14 wk after initial bacterial colonization, and native IEC from large intestine (cecum + colon) were isolated, as described in Materials and Methods. IEC from germfree mice were used as controls. Total protein was extracted, and 50 μg of protein was subjected to SDS-PAGE, followed by phospho-RelA, RelA, TLR2, and β-actin immunoblotting using the ECL technique. Gels represent the combined protein samples from each group.

FIGURE 2.

Differential RelA phosphorylation and TLR2 protein expression in native IEC from wild-type vs IL-10−/− mice. Germfree WT and IL-10−/− mice were colonized with E. faecalis at 12–16 wk of age. Mice (n = 3–5) were killed at days 3, 7, 14, and 28 as well as 14 wk after initial bacterial colonization, and native IEC from large intestine (cecum + colon) were isolated, as described in Materials and Methods. IEC from germfree mice were used as controls. Total protein was extracted, and 50 μg of protein was subjected to SDS-PAGE, followed by phospho-RelA, RelA, TLR2, and β-actin immunoblotting using the ECL technique. Gels represent the combined protein samples from each group.

Close modal

We next asked the question as to whether E. faecalis can directly induce RelA phosphorylation and NF-κB signal transduction in IEC lines. We first stimulated Mode-K and CMT-93 mouse epithelial cell lines with E. faecalis (moi of 100) for 0–4 h. Interestingly, E. faecalis transiently induced phospho-RelA in both cell lines (Fig. 3,A). To further dissect the signal transduction pathway of Gram-positive E. faecalis in IEC, we measured phospho-RelA, phospho-IκBα, total IκBα, and β-actin in Mode-K cells. As shown in Fig. 3 B, E. faecalis (moi of 100) induced phosphorylation of the cytoplasmic NF-κB inhibitor IκBα after 10–20 min of stimulation, followed by the induction of phospho-RelA after 30 min and complete IκBα degradation after 1 h of stimulation. This clearly demonstrates the capability of E. faecalis to activate the IκB/NF-κB complex in Mode-K cells.

FIGURE 3.

A and B, E. faecalis triggers RelA phosphorylation and IκB/NF-κB activation in CMT-93 and Mode-K mouse epithelial cell lines. Mode-K or CMT-93 cell lines were stimulated with E. faecalis for various times at an moi of 100. Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by immunoblotting using the ECL technique with Abs for phospho-RelA, RelA, phospho-IκBα, IκBα, and β-actin. Where indicated, the cells were pretreated for 1 h with the proteasome inhibitor MG132 (20 μM). Representative gels from at least three different experiments are shown.

FIGURE 3.

A and B, E. faecalis triggers RelA phosphorylation and IκB/NF-κB activation in CMT-93 and Mode-K mouse epithelial cell lines. Mode-K or CMT-93 cell lines were stimulated with E. faecalis for various times at an moi of 100. Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by immunoblotting using the ECL technique with Abs for phospho-RelA, RelA, phospho-IκBα, IκBα, and β-actin. Where indicated, the cells were pretreated for 1 h with the proteasome inhibitor MG132 (20 μM). Representative gels from at least three different experiments are shown.

Close modal

We next sought to examine E. faecalis-induced expression of IL-6 and IP-10 genes in Mode-K cells after the stimulation with E. faecalis (moi of 100) for 0–48 h. We measured IL-6 and IP-10 mRNA expression using real-time PCR and ELISA techniques. Fig. 4,A shows that E. faecalis transiently induced IL-6 and IP-10 mRNA expression with maximal induction (20- to 40-fold increase) after 12 h of bacterial stimulation (100 moi). Accordingly, E. faecalis-induced IL-6 and IP-10 protein secretion was maximal at 12 h of bacterial stimulation (Fig. 4 B).

FIGURE 4.

A and B, E. faecalis triggers IL-6 and IP-10 gene expression in Mode-K mouse epithelial cells. Mode-K or CMT-93 cell lines were stimulated with E. faecalis for various times at an moi of 100. A, Total RNA from Mode-K cells was extracted and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine IL-6, IP-10, and GAPDH. The induction of IL-6 and IP-10 mRNA in Mode-K cells was calculated relative to germfree controls (fold increase) using the crossing point of the log-linear portion of the amplification curve after normalization with GAPDH. B, IL-6 and IP-10 protein were measured in the spent culture supernatant from stimulated Mode-K cells using ELISA methods. The bars represent the combined mean value of duplicate samples measured in duplicates.

FIGURE 4.

A and B, E. faecalis triggers IL-6 and IP-10 gene expression in Mode-K mouse epithelial cells. Mode-K or CMT-93 cell lines were stimulated with E. faecalis for various times at an moi of 100. A, Total RNA from Mode-K cells was extracted and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine IL-6, IP-10, and GAPDH. The induction of IL-6 and IP-10 mRNA in Mode-K cells was calculated relative to germfree controls (fold increase) using the crossing point of the log-linear portion of the amplification curve after normalization with GAPDH. B, IL-6 and IP-10 protein were measured in the spent culture supernatant from stimulated Mode-K cells using ELISA methods. The bars represent the combined mean value of duplicate samples measured in duplicates.

Close modal

Mode-K cells were established from C3H/He (H-2K) mice (30). Later reports from Denning et al. (31) further specified the LPS-unresponsive mouse strain C3H/HeJ as the origin of Mode-K cells. To validate the characteristics of Mode-K cells with respect to their LPS responsiveness, we used normal Mode-K and TLR4/MD-2 stable transfected Mode-K cells. We stimulated the cells with E. faecalis (100 moi), LPS (10 μg/ml), and IL-1β (10 ng/ml) for 12 h and measured IL-6 and IP-10 protein secretion in the cell culture supernatants. Fig. 5,A shows that Mode-K cells were hyporesponsive to the stimulation with LPS, whereas E. faecalis and IL-1β significantly triggered IL-6 and IP-10 protein secretion. Interestingly, TLR4/MD-2-reconstituted Mode-K cells fully regained LPS responsiveness (Fig. 5 B), suggesting that Mode-K cells are indeed deficient in their TLR4 signaling cascade. These results validated Mode-K cells as a useful tool to characterize Gram-positive E. faecalis-induced signal transduction.

FIGURE 5.

A and B, Reconstitution of the TLR4/MD-2 complex reverted LPS hyporesponsiveness in Mode-K cells. Mode-K cells were transfected with a TLR4/MD-2 expression vector system, as described in Materials and Methods. Wild-type Mode-K cells (A) and stable transfected Mode-K TLR4/MD2 cells (B) were then stimulated with E. faecalis (moi 100), LPS (10 μg/ml), and IL-1β (20 ng/ml) for 12 h. IL-6 and IP-10 protein were measured in the cell culture supernatant using ELISA technique. The bars represent the combined mean value (±SD) of three experiments.

FIGURE 5.

A and B, Reconstitution of the TLR4/MD-2 complex reverted LPS hyporesponsiveness in Mode-K cells. Mode-K cells were transfected with a TLR4/MD-2 expression vector system, as described in Materials and Methods. Wild-type Mode-K cells (A) and stable transfected Mode-K TLR4/MD2 cells (B) were then stimulated with E. faecalis (moi 100), LPS (10 μg/ml), and IL-1β (20 ng/ml) for 12 h. IL-6 and IP-10 protein were measured in the cell culture supernatant using ELISA technique. The bars represent the combined mean value (±SD) of three experiments.

Close modal

We showed that the pattern recognition receptor TLR2 was differentially expressed in native IEC from E. faecalis-monoassociated WT and IL-10−/− mice (Fig. 2). To dissect the role of TLR2 in E. faecalis-induced signaling in IEC, we first stimulated WT and TLR2−/− MEF for 0–2 h with E. faecalis (moi of 100). As shown in Fig. 6,A, E. faecalis triggered RelA phosphorylation in WT, but not TLR2−/− MEF. Of note, IL-1β induced RelA phosphorylation in TLR2−/− MEF, confirming the specificity for the inhibition of E. faecalis-induced NF-κB signaling in TLR2−/− cells. In addition, E. faecalis-induced IL-6 secretion was completely abolished in TLR2−/− MEF compared with WT MEF (Fig. 6,B), demonstrating that nonpathogenic E. faecalis signals through the TL2R cascade to induce RelA phosphorylation and to activate NF-κB-dependent proinflammatory gene expression. Interestingly, IL-1β-induced IL-6 protein secretion was significantly increased in TLR2−/− MEF compared with WT cells (Fig. 6 B).

FIGURE 6.

A and B, E. faecalis signals through the TLR2 cascade to induce RelA phosphorylation and proinflammatory gene expression: proof of principle in wild-type and TLR2−/− MEF. WT and TLR2−/− MEF were stimulated with IL-1β (20 ng/ml) and E. faecalis at an moi of 100 for various times. A, Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by phospho-RelA and RelA immunoblotting using the ECL technique. B, IL-6 protein was measured in the cell culture supernatant after 12 h of stimulation using ELISA technique. The bars represent the combined mean value (±SD) of three experiments. The bars represent the combined mean value (±SD) of three experiments. ∗, p value <0.05; #, p value <0.05.

FIGURE 6.

A and B, E. faecalis signals through the TLR2 cascade to induce RelA phosphorylation and proinflammatory gene expression: proof of principle in wild-type and TLR2−/− MEF. WT and TLR2−/− MEF were stimulated with IL-1β (20 ng/ml) and E. faecalis at an moi of 100 for various times. A, Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by phospho-RelA and RelA immunoblotting using the ECL technique. B, IL-6 protein was measured in the cell culture supernatant after 12 h of stimulation using ELISA technique. The bars represent the combined mean value (±SD) of three experiments. The bars represent the combined mean value (±SD) of three experiments. ∗, p value <0.05; #, p value <0.05.

Close modal

To further address the question as to whether E. faecalis also signals through the TLR2 cascade in IEC, we established stable transfected Mode-K cells expressing mutant TLR2. We used a TLR2ΔTIR expression vector system and selected the transfected cells by adding nontoxic concentrations of puromycin to the medium. We then stimulated normal Mode-K and Mode-K TLR2ΔTIR cells with E. faecalis and IL-1β and measured phospho-RelA in total cell extracts as well as IL-6 protein in the cell culture supernatants. As shown in Fig. 7,A, E. faecalis-induced RelA phosphorylation was significantly reduced in Mode-K TLR2ΔTIR (lanes 5 and 6) compared with wild-type Mode-K cells (lanes 2 and 3). In contrast to TLR2−/− MEF (Fig. 6,A, lane 4), phospho-RelA background levels were increased in untreated TLR2ΔTIR Mode-K cells (lane 4). In addition and consistent with TLR2−/− MEF cell results, E. faecalis-induced IL-6 protein secretion was significantly reduced in Mode-K TLR2ΔTIR cells compared with wild-type Mode-K cells (Fig. 7 B). Of note, IL-1β stimulation revealed comparable responsiveness of Mode-K TLR2ΔTIR and normal Mode-K cells. The residual IL-6 response in E. faecalis-stimulated Mode-K TLR2ΔTIR cells may reflect the low transfection efficiency in IEC lines as well as an incomplete selection of TLR2ΔTIR-transfected cells. Nevertheless, these results clearly demonstrate that E. faecalis targets the TLR2 signal transduction pathway to induce proinflammatory gene expression in IEC. Similar results were obtained at the mRNA level for both MEF and Mode-K cells (data not shown).

FIGURE 7.

A and B, E. faecalis signals through the TLR2 cascade to induce RelA phosphorylation and proinflammatory gene expression in IEC. Mode-K cells were transfected with a TLR2ΔTIR expression vector system, as described in Materials and Methods. Wild-type Mode-K cells and stable transfected Mode-K TLR2ΔTIR cells were then stimulated with E. faecalis (moi 100) and IL-1β (20 ng/ml). A, Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by phospho-RelA and RelA immunoblotting using the ECL technique. B, IL-6 protein was measured in the cell culture supernatant after 12 h of stimulation using ELISA technique. The bars represent the combined mean value (±SD) of three experiments. The bars represent the combined mean value (±SD) of three experiments. ∗, p value <0.05.

FIGURE 7.

A and B, E. faecalis signals through the TLR2 cascade to induce RelA phosphorylation and proinflammatory gene expression in IEC. Mode-K cells were transfected with a TLR2ΔTIR expression vector system, as described in Materials and Methods. Wild-type Mode-K cells and stable transfected Mode-K TLR2ΔTIR cells were then stimulated with E. faecalis (moi 100) and IL-1β (20 ng/ml). A, Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by phospho-RelA and RelA immunoblotting using the ECL technique. B, IL-6 protein was measured in the cell culture supernatant after 12 h of stimulation using ELISA technique. The bars represent the combined mean value (±SD) of three experiments. The bars represent the combined mean value (±SD) of three experiments. ∗, p value <0.05.

Close modal

We next used adenoviral delivery of mutant IκB (Ad5IκBαAA), which cannot be phosphorylated or degraded, and dn IKKβ (Ad5dnIKKβ) to investigate the role of the IκB/NF-κB system to mediate bacteria-induced proinflammatory gene expression in IEC. We infected Mode-K cells with Ad5IκBαAA, Ad5dnIKKβ, and Ad5GFP as a viral control at an moi of 50 overnight and washed to remove the adenoviruses, and then stimulated Mode-K cells with E. faecalis (100 moi) for 12 h. Protein expression of HA- and FLAG-tagged mutant molecules was confirmed by Western blot analysis of cell lysates (Fig. 8,A). Interestingly, E. faecalis-induced IL-6 and IP-10 gene expression was significantly inhibited by the presence of either Ad5IκBαAA or Ad5dnIKKβ (Fig. 8 B), suggesting that E. faecalis triggers IκB/NF-κB activation to induce proinflammatory gene expression in IEC.

FIGURE 8.

A and B, E. faecalis triggers IL-6 and IP-10 gene expression in Mode-K cells through the IκB/NF-κB system. Mode-K cells were infected overnight with adenoviral Ad5IκBαAA and dn IKKβ (Ad5dnIKKβ) in serum-reduced cell culture medium. Expression of HA- and FLAG-tagged mutant molecules was controlled by Western blot analysis using mouse anti-HA and mouse anti-FLAG2 mAbs. Ad5GFP was used as viral negative control. Mode-K were then stimulated with E. faecalis (moi of 100). Total RNA was extracted after 12 h of stimulation and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine IL-6, IP-10, and GAPDH. The induction of IL-6 and IP-10 was calculated relative to untreated controls (fold increase) using the crossing point of the log-linear portion of the amplification curve after normalization with GAPDH. The bars represent the combined mean value (±SD) of three experiments. ∗, p value <0.05; ∗∗, p value <0.01.

FIGURE 8.

A and B, E. faecalis triggers IL-6 and IP-10 gene expression in Mode-K cells through the IκB/NF-κB system. Mode-K cells were infected overnight with adenoviral Ad5IκBαAA and dn IKKβ (Ad5dnIKKβ) in serum-reduced cell culture medium. Expression of HA- and FLAG-tagged mutant molecules was controlled by Western blot analysis using mouse anti-HA and mouse anti-FLAG2 mAbs. Ad5GFP was used as viral negative control. Mode-K were then stimulated with E. faecalis (moi of 100). Total RNA was extracted after 12 h of stimulation and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine IL-6, IP-10, and GAPDH. The induction of IL-6 and IP-10 was calculated relative to untreated controls (fold increase) using the crossing point of the log-linear portion of the amplification curve after normalization with GAPDH. The bars represent the combined mean value (±SD) of three experiments. ∗, p value <0.05; ∗∗, p value <0.01.

Close modal

It is well established that the immunosuppressive cytokine TGF-β activates the Smad pathway in numerous cell types (32, 33). Interestingly, phospho-Smad2 was induced in E. faecalis-monoassociated WT (at day 7), but not IL-10−/− mice (Fig. 8,A, upper blot). It appeared from these results that the absence of phospho-Smad2 expression was associated with persistent TLR2 expression and RelA phosphorylation in native IEC from IL-10−/− mice (Fig. 2), suggesting a potential role for the TGF-β/Smad signaling cascade in inhibiting TLR2-mediated NF-κB activity in native IEC. Of note, total Smad2/3 was equally expressed in IEC from WT and IL-10−/− mice. In addition, while the expression of the endogenous inhibitor of the TGF-β/Smad signaling cascade Smad 7 was induced in WT IEC at days 14 and 28 of E. faecalis monoassociation, IL-10−/− mice completely lack Smad7 expression in IEC (Fig. 9 A, lower blot).

FIGURE 9.

A and B, Differential TGF-β/Smad signaling in native IEC from E. faecalis-monoassociated wild-type vs IL-10−/− mice. Germfree WT and IL-10−/− mice were colonized with E. faecalis at 12–16 wk of age. Mice were killed at days 3, 7, 14, and 28 after initial bacterial colonization, and native IEC from large intestine (cecum + colon) were isolated, as described in Materials and Methods. IEC from germfree mice were used as controls. A, Total pooled protein was extracted, and 50 μg of protein was subjected to SDS-PAGE, followed by phospho-Smad2, Smad2/3, Smad7, and β-actin immunoblotting using the ECL technique. Gels represent the combined protein samples from each group. B, TGF-β1, TGF-βR2 mRNA expression was measured in native IEC and total mucosal tissue (cecum + colon) from WT and IL-10−/− mice using RT-PCR. Total RNA was extracted and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine TGF-β1, TGF-βR2, and GAPDH. PCR products were run on 2% agarose gels.

FIGURE 9.

A and B, Differential TGF-β/Smad signaling in native IEC from E. faecalis-monoassociated wild-type vs IL-10−/− mice. Germfree WT and IL-10−/− mice were colonized with E. faecalis at 12–16 wk of age. Mice were killed at days 3, 7, 14, and 28 after initial bacterial colonization, and native IEC from large intestine (cecum + colon) were isolated, as described in Materials and Methods. IEC from germfree mice were used as controls. A, Total pooled protein was extracted, and 50 μg of protein was subjected to SDS-PAGE, followed by phospho-Smad2, Smad2/3, Smad7, and β-actin immunoblotting using the ECL technique. Gels represent the combined protein samples from each group. B, TGF-β1, TGF-βR2 mRNA expression was measured in native IEC and total mucosal tissue (cecum + colon) from WT and IL-10−/− mice using RT-PCR. Total RNA was extracted and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine TGF-β1, TGF-βR2, and GAPDH. PCR products were run on 2% agarose gels.

Close modal

To further elucidate the mechanisms for the lack of Smad2 phosphorylation, we measured TGF-β1 and TGF-βR2 mRNA expression in native IEC and total mucosal tissue from WT and IL-10−/− mice. As shown in Fig. 9 B, TGF-β1 and TGF-βR2 mRNA expression were similar in native IEC as well as mucosal tissue from both WT and IL-10−/− mice. Of note, TGF-β1 and TGF-βR2 levels were constitutively expressed in colonic tissue from germfree and E. faecalis-monoassociated mice. Interestingly, E. faecalis monoassociation of WT and IL-10−/− mice induced TGF-β1 and to a lesser extent TGF-βR2 mRNA expression in colonic IEC after bacterial colonization.

To further verify whether the selective lack of Smad2 phosphorylation in IEC from IL-10−/− is attributed to intrinsic defects of TGF-β/Smad signaling in IEC, we stimulated primary IEC from germfree WT and IL-10−/− mice with TGF-β1 ex vivo. As shown in Fig. 10,A, TGF-β1 triggered transient Smad2 phosphorylation in both WT and IL-10−/− IEC. These results confirm the principal capability of TGF-β1 to trigger Smad signaling in primary IEC from IL-10−/− mice. Consistent with the transient Smad2 phosphorylation (day 7) and expression of the endogenous inhibitor of the TGF-β/Smad signaling cascade Smad7 (days 14 and 28) in native wild-type IEC (Fig. 9), we now demonstrate in Mode-K cells that adenoviral delivery of Smad7 blocked TGF-β1-induced Smad2 phosphorylation (Fig. 10 B).

FIGURE 10.

A and B, Ad5Smad7 inhibits TGF-β-mediated Smad2 phosphorylation in IEC. Primary IEC from the large intestine (cecum + colon) of germfree 129 SvEv WT and IL-10−/− mice were isolated and stimulated ex vivo with TGF-β1 (20 ng/ml) for 0–2 h (A). Mode-K cells were infected for 24 h with adenoviral Ad5Smad7 in serum-reduced cell culture medium. Mode-K cells were then stimulated with TGF-β1 (20 ng/ml) for 0–2 h in the presence and absence of Ad5Smad7 (B). Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by phospho-Smad2, Smad2/3, Smad7, and β-actin immunoblotting using the ECL technique.

FIGURE 10.

A and B, Ad5Smad7 inhibits TGF-β-mediated Smad2 phosphorylation in IEC. Primary IEC from the large intestine (cecum + colon) of germfree 129 SvEv WT and IL-10−/− mice were isolated and stimulated ex vivo with TGF-β1 (20 ng/ml) for 0–2 h (A). Mode-K cells were infected for 24 h with adenoviral Ad5Smad7 in serum-reduced cell culture medium. Mode-K cells were then stimulated with TGF-β1 (20 ng/ml) for 0–2 h in the presence and absence of Ad5Smad7 (B). Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by phospho-Smad2, Smad2/3, Smad7, and β-actin immunoblotting using the ECL technique.

Close modal

We showed that IL-10−/− mice compared with WT mice lack TGF-β/Smad signaling and fail to inhibit TLR2-mediated NF-κB activity in native IEC after initial colonization with nonpathogenic Gram-positive E. faecalis, well before the onset of histologic colitis. We next sought to investigate the effect of TGF-β1 on E. faecalis-induced RelA phosphorylation in Mode-K cells. Fig. 11,A shows that TGF-β1 triggers transient Smad2 phosphorylation after 1 h of stimulation of Mode-K (lower blot, lane 5), followed by the down-regulation of E. faecalis-induced phospho-RelA after 2 h of stimulation (upper blot, lane 6). These results were consistent with the findings in vivo, demonstrating that TGF-β1 inhibited E. faecalis-induced RelA phosphorylation. To provide direct evidence that TGF-β1 inhibits NF-κB-dependent IL-6 and IP-10 gene expression, we measured phospho-RelA DNA-binding activity on native IL-6 promoter sites using ChIP analysis. As shown in Fig. 11,B, TGF-β1 inhibited E. faecalis-induced phospho-RelA binding to IL-6 promoter binding sites, and accordingly, TGF-β1 inhibited E. faecalis-induced IL-6 and IP-10 gene expression in Mode-K cells (Fig. 11 C).

FIGURE 11.

A–C, TGF-β inhibits RelA phosphorylation and NF-κB-dependent proinflammatory gene expression in E. faecalis-stimulated Mode-K cells. Mode-K cells were stimulated with E. faecalis (moi of 100) in the absence or presence of TGF-β1 (20 ng/ml). A, Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by phospho-RelA, RelA, phospho-Smad2, and Smad2/3 immunoblotting using the ECL technique. Representative results from two to three independent experiments are shown. B, Phospho-RelA DNA binding to IL-6 gene promoters was measured using ChIP analysis, as described in Materials and Methods. Input control shows equal immunoprecipitation using anti-phospho-RelA Ab. C, Mode-K cells were stimulated for 12 h. Total RNA was extracted and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine IL-6, IP-10, and GAPDH. The induction of IL-6 and IP-10 mRNA was calculated relative to untreated controls (fold increase) using the crossing point of the log-linear portion of the amplification curve after normalization with GAPDH. The bars represent the combined mean value (±SD) of three experiments. ∗, p value <0.05.

FIGURE 11.

A–C, TGF-β inhibits RelA phosphorylation and NF-κB-dependent proinflammatory gene expression in E. faecalis-stimulated Mode-K cells. Mode-K cells were stimulated with E. faecalis (moi of 100) in the absence or presence of TGF-β1 (20 ng/ml). A, Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by phospho-RelA, RelA, phospho-Smad2, and Smad2/3 immunoblotting using the ECL technique. Representative results from two to three independent experiments are shown. B, Phospho-RelA DNA binding to IL-6 gene promoters was measured using ChIP analysis, as described in Materials and Methods. Input control shows equal immunoprecipitation using anti-phospho-RelA Ab. C, Mode-K cells were stimulated for 12 h. Total RNA was extracted and reverse transcribed, and real-time PCR was performed using the Light Cycler system with specific primers for murine IL-6, IP-10, and GAPDH. The induction of IL-6 and IP-10 mRNA was calculated relative to untreated controls (fold increase) using the crossing point of the log-linear portion of the amplification curve after normalization with GAPDH. The bars represent the combined mean value (±SD) of three experiments. ∗, p value <0.05.

Close modal

We next sought to investigate the role of TGF-β1 on TLR2 protein expression in Mode-K cells. We first stimulated Mode-K cells with TGF-β1 alone as well as in combination with E. faecalis and measured TLR2 protein expression after 1 and 2 h of stimulation. Interestingly, while the treatment of Mode-K cells with E. faecalis (Fig. 12,A, lanes 4 and 5) and TGF-β1 (Fig. 12,A, lanes 2 and 3) alone did not affect TLR2 protein expression in Mode-K cells (Fig. 12,A, lane 1), the combined stimulation with E. faecalis and TGF-β1 completely abolished TLR2 protein expression (Fig. 12,A, lanes 6 and 7). To further characterize the role of the TGF-β/Smad signaling pathway in the regulation of TLR2 expression in E. faecalis-stimulated Mode-K cells, we used the adenoviral vector Ad5Smad7 to block the TGF-β/Smad pathway (see Fig. 10,B) and MG132 to inhibit proteasome activity (see Fig. 3,B) in Mode-K cells. As shown in Fig. 12 B, the presence of Ad5Smad7 (lanes 3 and 4), but not MG132 (lanes 5 and 6), inhibited TGF-β-induced TLR2 expression in E. faecalis-stimulated Mode-K cells.

FIGURE 12.

A and B, TGF-β/Smad signaling induces TLR2 protein degradation in E. faecalis-stimulated Mode-K cells. Mode-K cells were stimulated with E. faecalis (moi of 100) for 0–2 h in the absence or presence of TGF-β1 (20 ng/ml) (A). Mode-K cells were infected for 24 h with Ad5Smad7 in serum-reduced cell culture medium. Where indicated, Mode-K cells were pretreated with the proteasome inhibitor MG132 (20 μM). Ad5Smad7-infected Mode-K and MG132-pretreated Mode-K cells were stimulated with E. faecalis (moi 100) for 0–2 h (B). Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by TLR2 and β-actin by immunoblotting using the ECL technique. Representative Western blots from two different experiments are shown.

FIGURE 12.

A and B, TGF-β/Smad signaling induces TLR2 protein degradation in E. faecalis-stimulated Mode-K cells. Mode-K cells were stimulated with E. faecalis (moi of 100) for 0–2 h in the absence or presence of TGF-β1 (20 ng/ml) (A). Mode-K cells were infected for 24 h with Ad5Smad7 in serum-reduced cell culture medium. Where indicated, Mode-K cells were pretreated with the proteasome inhibitor MG132 (20 μM). Ad5Smad7-infected Mode-K and MG132-pretreated Mode-K cells were stimulated with E. faecalis (moi 100) for 0–2 h (B). Total protein was extracted, and 20 μg of protein was subjected to SDS-PAGE, followed by TLR2 and β-actin by immunoblotting using the ECL technique. Representative Western blots from two different experiments are shown.

Close modal

In conclusion, we showed that TGF-β/Smad signaling was associated with the loss of TLR2 protein expression and the inhibition of bacteria-induced RelA phosphorylation and NF-κB-dependent gene expression in native and IEC lines.

In this study, we demonstrate that E. faecalis-monoassociated WT and IL-10−/− mice differentially induce NF-κB and TGF-β/Smad signaling in the intestinal epithelium at early stages of bacterial colonization. Consistent with the finding that E. faecalis triggers RelA phosphorylation and NF-κB-dependent IL-6 and IP-10 gene expression through the TLR2 cascade, TLR2 protein expression was absent after day 7 from E. faecalis-monoassociated WT mice, but persisted in IEC from IL-10−/− mice. Most relevant for the mechanistic basis of altered NF-κB/TLR signaling in IEC from WT and IL-10−/− mice is the finding that TGF-β1 triggers TLR2 protein degradation and E. faecalis-induced NF-κB-dependent gene expression in epithelial cell lines. It was previously shown that E. faecalis-monoassociated IL-10−/−, but not WT mice develop experimental colitis after 12–16 wk after initial bacterial colonization (12, 13), demonstrating the ability of this nonpathogenic Gram-positive bacterial strain to induce immune-mediated colitis in a susceptible host. Despite the findings that E. faecalis monoassociation progressively induced enteric bacterial Ag-specific CD4+ T cell activation and IFN-γ production in IL-10−/− mice, the underlying mechanisms responsible for initiating intestinal inflammation are essentially unknown (13).

E. faecalis induces persistent RelA phosphorylation in native IEC from monoassociated IL-10−/− mice at days 3, 7, 14, and 28 after bacterial colonization, while colonized WT mice transiently trigger phospho-RelA at day 7. Consistent with the development of histopathology, IP-10 gene expression was transient after 1 wk of E. faecalis monoassociation in IEC from WT, but persisted after 14 wk of bacterial colonization in IL-10−/− mice. This may suggest an intrinsic defect of IL-10−/− mice to control transient induction of NF-κB activity in the intestinal epithelium of WT mice following bacterial colonization, supporting the possibility that bacteria-epithelial cell interactions contribute to the induction of immunopathology, leading to chronic intestinal inflammation in the genetically susceptible host colonized with enteric bacteria.

Increased NF-κB activation has been well documented in IEC and lamina propria cells of IBD patients with active disease (34, 35, 36), and accordingly, pharmacological NF-κB blockade may become potentially important in the treatment of chronic intestinal inflammation (37, 38). In addition, NF-κB RelA was strongly activated in the intestinal mucosa of IL-10−/− and trinitrobenzene sulfonic acid (TNBS)-treated mice (39). Most importantly, local administration of antisense RelA oligonucleotides abrogated clinical and histological signs of TNBS-induced experimental colitis, suggesting a mechanistic role for sustained NF-κB activity in the pathogenesis of chronic mucosal inflammation (39). In contrast, the selective inhibition of NF-κB activity in enterocytes of IKKβ gene-deficient mice sensitized these animals to ischemia-reperfusion-induced apoptosis in IEC, which was associated with a loss of mucosal integrity (40). In addition, blocking NF-κB activity with pharmacological inhibitors during the resolution phase of carrageenan-induced acute inflammation is deleterious to the host (41), suggesting dual functions of activated NF-κB, including protective and detrimental mechanisms during the course of inflammation.

Although many pathways are involved in the regulation of innate and/or adaptive immunity in the intestine, the immunoregulatory cytokines TGF-β and IL-10 are of high relevance to IBD, as elegantly demonstrated in TGF-β1- and IL-10-deficient mice that both spontaneously develop colitis (42). Accordingly, TGF-β1 overexpression in lamina propria immune cells inhibited Th1-mediated experimental TNBS-induced colitis (6, 43), and mucosal delivery of IL-10-secreting lactococci abrogated experimental colitis in IL-10−/− mice (44). The importance of TGF-β signaling in maintaining epithelial cell homeostasis was demonstrated in tissue-specific transgenic mice expressing a dn TGF-β receptor in IEC (TGF-R2). Molecular blockade of TGF-β1 signaling triggered colitis in TGF-R2 transgenic mice under conventional conditions and increased the susceptibility of these mice to dextran-sodium-sulfate-induced colitis (45).

Key features of our present study are the observations that the lack of Smad2 phosphorylation in IEC from E. faecalis-monoassociated IL-10−/− mice is associated with persistent TLR2 and IP-10 expression. Although the endogenous inhibitor of the TGF-β signaling cascade Smad7 is implicated in the development of chronic intestinal inflammation (46, 47) and blocked TGF-β-induced Smad2 phosphorylation in Mode-K (Ad5Smad7 infected) and other epithelial cell lines (48), Smad7 was not expressed in IL-10−/− IEC. In addition, the expression of TGF-β1 and the TGF-β-binding receptor TGF-βR2 mRNA was similar in WT and IL-10−/− IEC and, most importantly, TGF-β1 triggered Smad2 phosphorylation in ex vivo cultures from WT as well as IL-10−/− IEC. It appears from these findings that the differential Smad2 phosphorylation in IEC from E. faecalis-monoassociated WT vs IL-10−/− mice is not due to an intrinsic defect of responsiveness of the TGF-β/Smad cascade in IEC. Functional secretion of biologically active TGF-β requires posttranscriptional modifications and may be altered in IL-10−/− mice. In addition to the enormously difficult experimental conditions to measure secreted bioactive TGF-β from intestinal tissue and/or cells, functional TGF-β is also cell membrane bound, suggesting that the physiological inhibitory effects of TGF-β could also be locally mediated on the basis of cell-to-cell contact (49). Interestingly, Strober and colleagues (50, 51) demonstrated that the protective mechanisms of IL-10 in TNBS-induced experimental colitis are indirectly mediated through its inductive effect on TGF-β secretion in lamina propria T cells. Most importantly, CD45RBlow T cells from IL-10−/− mice fail to confer protection in an adoptive transfer model in which CD45RBhigh T cells trigger Th1-mediated murine experimental colitis in SCID mice (52, 53), supporting the possibility that IL-10 could be important in triggering TGF-β-producing T cells in the normal intestinal mucosa. An attractive hypothesis is that transient induction of NF-κB activity in epithelial cells triggers biologically active IL-10-mediated TGF-β responses in the lamina propria or the epithelium.

The intestinal epithelium must adapt to a constantly changing luminal environment by processing different biological information through multiple signaling cascades that target a defined set of genes, to provide an adequate effector response. We showed that Gram-positive E. faecalis can directly induce RelA phosphorylation in IEC lines and native IEC from WT mice, demonstrating the physiological relevance of E. faecalis in targeting the NF-κB cascade. Under normal conditions, E. faecalis-induced RelA phosphorylation and proinflammatory gene expression were transient, suggesting the presence of intrinsic mechanisms that antagonize E. faecalis-induced NF-κB activity in IEC. These regulatory mechanisms may be an intrinsic part of the NF-κB pathway such as Toll-interacting protein or IL-1R-associated kinase-M operating in a negative feedback loop fashion and/or alternatively be induced by independent signaling cascades that cross talk with NF-κB cascade, including TGF-β/Smad and 15-deoxy-Δ12,14 PGJ2-ERK signaling (19, 54, 55, 56).

We performed a series of experiments to demonstrate that E. faecalis signals through TLR2 to trigger NF-κB activation, and, most importantly, to characterize the mechanistic basis for the inhibitory effects of TGF-β1. First, we show that E. faecalis induced IκBα and RelA phosphorylation, followed by complete IκBα degradation in Mode-K cells. Second and most relevant to elucidate the mechanisms of bacterial signaling in IEC, adenoviral delivery of IκBα superrepressor and dn IKK-β inhibited IL-6, IP-10, and TLR2 gene expression in Mode-K cells. This clearly demonstrates the pivotal role of the IκB/NF-κB system in E. faecalis-induced signal transduction in IEC. Third, functional evidence for the importance of TLR2 in E. faecalis signaling was achieved by using TLR2−/− MEF and TLR2 mutant Mode-K cells. We could demonstrate that E. faecalis-induced RelA phosphorylation as well as proinflammatory gene expression were abrogated in TLR2−/− MEF as well as Mode-K TLR2ΔTIR. It has been recently reported that the extracellular domain of TLR2 contains multiple binding domains for different Gram-positive bacteria-derived ligands (57), which may explain the differential ability of complex Gram-positive bacteria to induce TLR2 signaling and NF-κB activation in IEC (58). Finally and most relevant for a possible explanation of altered bacterial-induced TLR/NF-κB signaling in IL-10−/− IEC, we showed that TGF-β1 down-regulates TLR2 protein expression in IEC, followed by the inhibition of E. faecalis-induced RelA phosphorylation and proinflammatory gene expression. Interestingly, TGF-β-mediated degradation of TLR2 was induced in the presence, but not in the absence of E. faecalis stimulation. Similarly, Matsumura et al. (59) demonstrated that TGF-β up-regulated TLR2 expression in hepatocytes, but inhibited TLR2 expression induced by IL-1β. Together with our observation that the presence of Ad5Smad7, but not the pharmacological proteasome inhibitor MG132, blocked TGF-β-mediated TLR2 degradation, we may conclude from our findings that the TGF-β/Smad signaling cascade induces TLR2 degradation in E. faecalis-stimulated IEC most likely through a proteasome-independent pathway. Although TLR expression has been shown to be regulated through distinct E3 ubiquitin-protein ligases that selectively direct the proteasomal pathway to TLRs for proteolytic degradation (60), the recruitment of activated TLR1, 2, and 4 to endosomal/lysosomal degradation upon ligand-driven TLR activation may be an interesting new target of regulation (61, 62)

In conclusion, we demonstrate that Gram-positive nonpathogenic E. faecalis transiently induced TLR2-mediated RelA phosphorylation and NF-κB-dependent gene expression in IEC from WT mice, but persistent activation of the TLR/NF-κB pathway in IL-10-deficient mice at early stages of bacterial colonization preceding any histological evidence of colitis. Our results suggest the possibility that E. faecalis triggers persistently active TLR/NF-κB signaling in epithelial cells of IL-10−/− mice in the absence of the protective TGF-β/Smad cascade, which may lead indeed to the development of clinical and histological signs of intestinal inflammation. These results support the concept for an interrelated role of IL-10 and TGF-β1 in maintaining epithelial cell homeostasis to commensal enteric bacteria.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by Die Deutsche Forschungsgemeinschaft Grant HA 3148/1-2 (to D.H.); National Institutes of Health RO1 Grants RO1 DK53347 and P30 DK34987 (to R.B.S.); and the Crohn’s and Colitis Foundation of America (to S.C.K.).

4

Abbreviations used in this paper: IBD, inflammatory bowel disease; ChIP, chromatin immunoprecipitation; dn, dominant negative; HA, hemagglutinin; IEC, intestinal epithelial cell; IKK, I-κB kinase; IP-10, IFN-γ-inducible protein-10; MEF, myoembryogenic fibroblast; moi, multiplicity of infection; TIR, Toll/IL-1R; TNBS, trinitrobenzene sulfonic acid; WT, wild type.

1
Podolsky, D. K..
2002
. Inflammatory bowel disease.
N. Engl. J. Med.
347
:
417
.
2
Bouma, G., W. Strober.
2003
. The immunological and genetic basis of inflammatory bowel disease.
Nat. Rev. Immunol.
3
:
521
.
3
Sartor, R. B..
1997
. The influence of normal microbial flora on the development of chronic mucosal inflammation.
Res. Immunol.
148
:
567
.
4
Haller, D., C. Jobin.
2004
. Interaction between resident luminal bacteria and the host: can a healthy relationship turn sour?.
J. Pediatr. Gastroenterol. Nutr.
38
:
123
.
5
Duchmann, R., I. Kaiser, E. Hermann, W. Mayet, K. Ewe, K. H. Meyer zum Buschenfelde.
1995
. Tolerance exists towards resident intestinal flora but is broken in active inflammatory bowel disease (IBD).
Clin. Exp. Immunol.
102
:
448
.
6
Duchmann, R., E. Schmitt, P. Knolle, K. H. Meyer zum Buschenfelde, M. Neurath.
1996
. Tolerance towards resident intestinal flora in mice is abrogated in experimental colitis and restored by treatment with interleukin-10 or antibodies to interleukin-12.
Eur. J. Immunol.
26
:
934
.
7
Duchmann, R., E. May, M. Heike, P. Knolle, M. Neurath, K. H. Meyer zum Buschenfelde.
1999
. T cell specificity and cross reactivity towards enterobacteria, Bacteroides, bifidobacterium, and antigens from resident intestinal flora in humans.
Gut
44
:
812
.
8
Harper, P. H., E. C. Lee, M. G. Kettlewell, M. K. Bennett, D. P. Jewell.
1985
. Role of the faecal stream in the maintenance of Crohn’s colitis.
Gut
26
:
279
.
9
Rutgeerts, P., K. Goboes, M. Peeters, M. Hiele, F. Penninckx, R. Aerts, R. Kerremans, G. Vantrappen.
1991
. Effect of faecal stream diversion on recurrence of Crohn’s disease in the neoterminal ileum.
Lancet
338
:
771
.
10
D’Haens, G. R., K. Geboes, M. Peeters, F. Baert, F. Penninckx, P. Rutgeerts.
1998
. Early lesions of recurrent Crohn’s disease caused by infusion of intestinal contents in excluded ileum.
Gastroenterology
114
:
262
.
11
Strober, W., I. J. Fuss, R. S. Blumberg.
2002
. The immunology of mucosal models of inflammation.
Annu. Rev. Immunol.
20
:
495
.
12
Balish, E., T. Warner.
2002
. Enterococcus faecalis induces inflammatory bowel disease in interleukin-10 knockout mice.
Am. J. Pathol.
160
:
2253
.
13
Kim, S., S. L. Tonkonogy, C. A. Albright, J. Tsang, E. J. Balish, J. Braun, M. M. Huycke, R. B. Sartor.
2003
. Variable phenotypes of enterocolitis in IL-10 deficient mice manoassociated with two different commensal bacteria.
Gastroenterology
124
:
A1106
.
14
Jobin, C., R. B. Sartor.
2000
. The IκB/NF-κB system: a key determinant of mucosal inflammation and protection.
Am. J. Physiol. Cell Physiol.
278
:
C451
.
15
Hershberg, R. M., P. E. Framson, D. H. Cho, L. Y. Lee, S. Kovats, J. Beitz, J. S. Blum, G. T. Nepom.
1997
. Intestinal epithelial cells use two distinct pathways for HLA class II antigen processing.
J. Clin. Invest.
100
:
204
.
16
Kagnoff, M. F., L. Eckmann.
1997
. Epithelial cells as sensors for microbial infection.
J. Clin. Invest.
100
:
6
.
17
Blumberg, R. S., C. Terhorst, P. Bleicher, F. V. McDermott, C. H. Allan, S. B. Landau, J. S. Trier, S. P. Balk.
1991
. Expression of a nonpolymorphic MHC class I-like molecule, CD1D, by human intestinal epithelial cells.
J. Immunol.
147
:
2518
.
18
Pahl, H. L..
1999
. Activators and target genes of Rel/NF-κB transcription factors.
Oncogene
18
:
5853
.
19
Haller, D., L. Holt, S. C. Kim, R. F. Schwabe, R. B. Sartor, C. Jobin.
2003
. Transforming growth factor-β1 inhibits non-pathogenic Gram negative bacteria-induced NF-κB recruitment to the interleukin-6 gene promoter in intestinal epithelial cells through modulation of histone acetylation.
J. Biol. Chem.
278
:
23851
.
20
Haller, D., M. P. Russo, R. B. Sartor, C. Jobin.
2002
. IKKβ and phosphatidylinositol 3-kinase/Akt participate in non-pathogenic Gram-negative enteric bacteria-induced RelA phosphorylation and NF-κB activation in both primary and intestinal epithelial cell lines.
J. Biol. Chem.
277
:
38168
.
21
Hornef, M. W., T. Frisan, A. Vandewalle, S. Normark, A. Richter-Dahlfors.
2002
. Toll-like receptor 4 resides in the Golgi apparatus and colocalizes with internalized lipopolysaccharide in intestinal epithelial cells.
J. Exp. Med.
195
:
559
.
22
Backhed, F., M. Hornef.
2003
. Toll-like receptor 4-mediated signaling by epithelial surfaces: necessity or threat?.
Microbes Infect.
5
:
951
.
23
Dunne, A., and L. A. O’Neill. 2003. The interleukin-1 receptor/Toll-like receptor superfamily: signal transduction during inflammation and host defense. Sci. STKE 2003:re3.
24
Takeda, K., T. Kaisho, S. Akira.
2003
. Toll-like receptors.
Annu. Rev. Immunol.
21
:
335
.
25
Zhang, G., S. Ghosh.
2001
. Toll-like receptor-mediated NF-κB activation: a phylogenetically conserved paradigm in innate immunity.
J. Clin. Invest.
107
:
13
.
26
Hausmann, M., S. Kiessling, S. Mestermann, G. Webb, T. Spottl, T. Andus, J. Scholmerich, H. Herfarth, K. Ray, W. Falk, G. Rogler.
2002
. Toll-like receptors 2 and 4 are up-regulated during intestinal inflammation.
Gastroenterology
122
:
1987
.
27
Cario, E., D. K. Podolsky.
2000
. Differential alteration in intestinal epithelial cell expression of Toll-like receptor 3 (TLR3) and TLR4 in inflammatory bowel disease.
Infect. Immun.
68
:
7010
.
28
Rath, H. C., H. H. Herfarth, J. S. Ikeda, W. B. Grenther, T. E. Hamm, Jr, E. Balish, J. D. Taurog, R. E. Hammer, K. H. Wilson, R. B. Sartor.
1996
. Normal luminal bacteria, especially Bacteroides species, mediate chronic colitis, gastritis, and arthritis in HLA-B27/human β2 microglobulin transgenic rats.
J. Clin. Invest.
98
:
945
.
29
Pfaffl, M. W..
2001
. A new mathematical model for relative quantification in real-time RT-PCR.
Nucleic Acids Res.
29
:
e45
.
30
Vidal, K., I. Grosjean, J. P. Evillard, C. Gespach, D. Kaiserlian.
1993
. Immortalization of mouse intestinal epithelial cells by the SV40-large T gene: phenotypic and immune characterization of the MODE-K cell line.
J. Immunol. Methods
166
:
63
.
31
Denning, T. L., N. A. Campbell, F. Song, R. P. Garofalo, G. R. Klimpel, V. E. Reyes, P. B. Ernst.
2000
. Expression of IL-10 receptors on epithelial cells from the murine small and large intestine.
Int. Immunol.
12
:
133
.
32
Moore, K. W., R. de Waal Malefyt, R. L. Coffman, A. O’Garra.
2001
. Interleukin-10 and the interleukin-10 receptor.
Annu. Rev. Immunol.
19
:
683
.
33
Shi, Y., J. Massague.
2003
. Mechanisms of TGF-β signaling from cell membrane to the nucleus.
Cell
113
:
685
.
34
Rogler, G., K. Brand, D. Vogl, S. Page, R. Hofmeister, T. Andus, R. Knuechel, P. A. Baeuerle, J. Scholmerich, V. Gross.
1998
. Nuclear factor κB is activated in macrophages and epithelial cells of inflamed intestinal mucosa.
Gastroenterology
115
:
357
.
35
Schreiber, S., S. Nikolaus, J. Hampe.
1998
. Activation of nuclear factor κB inflammatory bowel disease.
Gut
42
:
477
.
36
Neurath, M. A., C. Becker, K. Barbulescu.
1998
. Role of NF-κB in immune and inflammatory response in the gut.
Gut
43
:
856
.
37
Ardite, E., J. Panes, M. Miranda, A. Salas, J. I. Elizalde, M. Sans, Y. Arce, J. M. Bordas, J. C. Fernandez-Checa, J. M. Pique.
1998
. Effects of steroid treatment on activation of nuclear factor κB in patients with inflammatory bowel disease.
Br. J. Pharmacol.
124
:
431
.
38
Wahl, C., S. Liptay, G. Adler, R. M. Schmid.
1998
. Sulfasalazine: a potent and specific inhibitor of nuclear factor κB.
J. Clin. Invest.
101
:
1163
.
39
Neurath, M., S. Pettersson, K.-H. Meyer zum Buschenfelde, W. Strober.
1996
. Local administration of antisense phosphothioate oligonucleotides to the p65 subunit of NF-κB abrogates established experimental colitis in mice.
Nat. Med.
2
:
998
.
40
Chen, L. W., L. Egan, Z. W. Li, F. R. Greten, M. F. Kagnoff, M. Karin.
2003
. The two faces of IKK and NF-κB inhibition: prevention of systemic inflammation but increased local injury following intestinal ischemia-reperfusion.
Nat. Med.
9
:
575
.
41
Lawrence, T., D. W. Gilroy, P. R. Colville-Nash, D. A. Willoughby.
2001
. Possible new role for NF-κB in the resolution of inflammation.
Nat. Med.
7
:
1291
.
42
Kulkarni, A. B., J. M. Ward, L. Yaswen, C. L. Mackall, S. R. Bauer, C. G. Huh, R. E. Gress, S. Karlsson.
1995
. Transforming growth factor-β1 null mice: an animal model for inflammatory disorders.
Am. J. Pathol.
146
:
264
.
43
Kitani, A., I. J. Fuss, K. Nakamura, O. M. Schwartz, T. Usui, W. Strober.
2000
. Treatment of experimental (trinitrobenzene sulfonic acid) colitis by intranasal administration of transforming growth factor (TGF)-β1 plasmid: TGF-β1-mediated suppression of T helper cell type 1 response occurs by interleukin (IL)-10 induction and IL-12 receptor β2 chain down-regulation.
J. Exp. Med.
192
:
41
.
44
Steidler, L., W. Hans, L. Schotte, S. Neirynck, F. Obermeier, W. Falk, W. Fiers, E. Remaut.
2000
. Treatment of murine colitis by Lactococcus lactis secreting interleukin-10.
Science
289
:
1352
.
45
Hahm, K. B., Y. H. Im, T. W. Parks, S. H. Park, S. Markowitz, H. Y. Jung, J. Green, S. J. Kim.
2001
. Loss of transforming growth factor β signalling in the intestine contributes to tissue injury in inflammatory bowel disease.
Gut
49
:
190
.
46
Monteleone, G., A. Kumberova, N. M. Croft, C. McKenzie, H. W. Steer, T. T. MacDonald.
2001
. Blocking Smad7 restores TGF-β1 signaling in chronic inflammatory bowel disease.
J. Clin. Invest.
108
:
601
.
47
Monteleone, G., J. Mann, I. Monteleone, P. Vavassori, R. Bremner, M. Fantini, G. Del Vecchio Blanco, R. Tersigni, L. Alessandroni, D. Mann, et al
2003
. A failure of TGFβ1 negative regulation maintains sustained NF-κB activation in gut inflammation.
J. Biol. Chem.
4
:
4
.
48
Li, J. H., H. J. Zhu, X. R. Huang, K. N. Lai, R. J. Johnson, H. Y. Lan.
2002
. Smad7 inhibits fibrotic effect of TGF-β on renal tubular epithelial cells by blocking Smad2 activation.
J. Am. Soc. Nephrol.
13
:
1464
.
49
Nakamura, K., A. Kitani, W. Strober.
2001
. Cell contact-dependent immunosuppression by CD4+CD25+ regulatory T cells is mediated by cell surface-bound transforming growth factor β.
J. Exp. Med.
194
:
629
.
50
Fuss, I. J., M. Boirivant, B. Lacy, W. Strober.
2002
. The interrelated roles of TGF-β and IL-10 in the regulation of experimental colitis.
J. Immunol.
168
:
900
.
51
Neurath, M. F., I. Fuss, B. L. Kelsall, D. H. Presky, W. Waegell, W. Strober.
1996
. Experimental granulomatous colitis in mice is abrogated by induction of TGF-β-mediated oral tolerance.
J. Exp. Med.
183
:
2605
.
52
Asseman, C., S. Mauze, M. W. Leach, R. L. Coffman, F. Powrie.
1999
. An essential role for interleukin 10 in the function of regulatory T cells that inhibit intestinal inflammation.
J. Exp. Med.
190
:
995
.
53
Powrie, F., J. Carlino, M. W. Leach, S. Mauze, R. L. Coffman.
1996
. A critical role for transforming growth factor-β but not interleukin 4 in the suppression of T helper type 1-mediated colitis by CD45RBlow CD4+ T cells.
J. Exp. Med.
183
:
2669
.
54
Ruiz, P. A., S. C. Kim, R. Balfour Sartor, D. Haller.
2004
. 15-Deoxy-δ(12,14)-prostaglandin J2-mediated ERK signaling inhibits Gram negative bacteria-induced RelA phosphorylation and IL-6 gene expression in intestinal epithelial cells through modulation of protein phosphatase 2A activity.
J. Biol. Chem.
21
:
21
.
55
Otte, J. M., E. Cario, D. K. Podolsky.
2004
. Mechanisms of cross hyporesponsiveness to Toll-like receptor bacterial ligands in intestinal epithelial cells.
Gastroenterology
126
:
1054
.
56
Kobayashi, K., L. D. Hernandez, J. E. Galan, C. A. Janeway, Jr, R. Medzhitov, R. A. Flavell.
2002
. IRAK-M is a negative regulator of Toll-like receptor signaling.
Cell
110
:
191
.
57
Meng, G., A. Grabiec, M. Vallon, B. Ebe, S. Hampel, W. Bessler, H. Wagner, C. J. Kirschning.
2003
. Cellular recognition of tri-/di-palmitoylated peptides is independent from a domain encompassing the N-terminal seven leucine-rich repeat (LRR)/LRR-like motifs of TLR2.
J. Biol. Chem.
278
:
39822
.
58
Melmed, G., L. S. Thomas, N. Lee, S. Y. Tesfay, K. Lukasek, K. S. Michelsen, Y. Zhou, B. Hu, M. Arditi, M. T. Abreu.
2003
. Human intestinal epithelial cells are broadly unresponsive to Toll-like receptor 2-dependent bacterial ligands: implications for host-microbial interactions in the gut.
J. Immunol.
170
:
1406
.
59
Matsumura, T., H. Hayashi, T. Takii, C. F. Thorn, A. S. Whitehead, J. Inoue, K. Onozaki.
2004
. TGF-β down-regulates IL-1α-induced TLR2 expression in murine hepatocytes.
J. Leukocyte Biol.
75
:
1056
.
60
Chuang, T. H., R. J. Ulevitch.
2004
. Triad3A, an E3 ubiquitin-protein ligase regulating Toll-like receptors.
Nat. Immunol.
5
:
495
.
61
Nomura, F., S. Akashi, Y. Sakao, S. Sato, T. Kawai, M. Matsumoto, K. Nakanishi, M. Kimoto, K. Miyake, K. Takeda, S. Akira.
2000
. Cutting edge: endotoxin tolerance in mouse peritoneal macrophages correlates with down-regulation of surface Toll-like receptor 4 expression.
J. Immunol.
164
:
3476
.
62
Heil, F., P. Ahmad-Nejad, H. Hemmi, H. Hochrein, F. Ampenberger, T. Gellert, H. Dietrich, G. Lipford, K. Takeda, S. Akira, et al
2003
. The Toll-like receptor 7 (TLR7)-specific stimulus loxoribine uncovers a strong relationship within the TLR7, 8 and 9 subfamily.
Eur. J. Immunol.
33
:
2987
.