Supported by the National Institutes of Health (DK075386-0251, R03DK089010-01), the American Cancer Society (RSG-09-075-01-MBC), and the IDEAL award from New York State's Empire State Stem Cell Board (N09G-279) to Jun Sun and NIHR01 DK053347, P40 RR018603, P30 DK034987 and the Crohn's and Colitis foundation of America to R.Balfour Sartor.
Wnt2 inhibits enteric bacterial-induced inflammation in intestinal epithelial cells†
Article first published online: 14 JUN 2011
Copyright © 2011 Crohn's & Colitis Foundation of America, Inc.
Inflammatory Bowel Diseases
Volume 18, Issue 3, pages 418–429, March 2012
How to Cite
Liu, X., Lu, R., Wu, S., Zhang, Y.-g., Xia, Y., Sartor, R. B. and Sun, J. (2012), Wnt2 inhibits enteric bacterial-induced inflammation in intestinal epithelial cells. Inflamm Bowel Dis, 18: 418–429. doi: 10.1002/ibd.21788
- Issue published online: 13 FEB 2012
- Article first published online: 14 JUN 2011
- Manuscript Accepted: 9 MAY 2011
- Manuscript Received: 2 MAY 2011
- gnotobiotic mice;
- E. coli;
Wnt signaling plays an essential role in gastrointestinal epithelial proliferation. Most investigations have focused on developmental and immune responses. Bacterial infection can be chronic and increases the risk of inflammatory bowel disease and colitis-associated cancer. However, we lack studies on how bacteria regulate Wnt proteins and how Wnts modulate the host responses to enteric bacteria. This study investigated the effects of Salmonella and Escherichia coli on Wnt2, one of the Wnt family members, in intestinal epithelia cells.
Using cultured epithelial cells, a Salmonella-colitis mouse model, and a gnotobiotic mouse model, we found that Wnt2 mRNA and protein expression levels were elevated after bacterial infection. Enteric bacteria regulate Wnt2 location in the intestine. Furthermore, we found that elevation of Wnt2 was a strategy for host defense by inhibiting cell apoptosis and inflammatory responses to infection.
Using Wnt2 siRNA analysis, we show enhanced inflammatory cytokine IL-8 in epithelial cells. Cells overexpressed Wnt2 had less bacterial-induced IL-8 secretion. AvrA is a bacterial protein that inhibits inflammation by stabilizing β-catenin, the downstream target of Wnt. We found that the stabilization of Wnt2 was regulated through ubiquitination. Moreover, the bacterial protein AvrA from Salmonella and E. coli stabilized Wnt2 protein expression in vivo. In an ex-germ-free system, E. coli F18 expressing AvrA increased Wnt2 expression and changed Wnt2 distribution in intestine.
Wnt2 contributes to host protection in response to enteric bacteria. Our findings thus reveal a previously undefined role of Wnt for host–pathogen interaction and inflammation. (Inflamm Bowel Dis 2012;)
The glycoprotein ligand Wnt proteins, including Wnt2, control cell fate during embryonic development and homeostasis in adult self-renewing tissues.1 Wnts are known to activate the β-catenin pathway, a regulator of intestinal epithelial proliferation and inflammation.2–7 Recently, several studies have implied the involvement of Wnt-Frizzled signaling in the activation of proinflammatory mediators in inflammatory disorders.8 Patients with inflammatory bowel disease (IBD) have significantly higher expression of Wnts compared to non-IBD patients.9 Moreover, the transmembrane Wnt receptors, Frizzled (Fz)3 and Fz4, exhibited significantly increased expression, whereas Fz1 and Fz5 exhibited significantly decreased expression in ulcerative colitis patients.9 A recent study indicated a reduced expression of the Wnt-signaling pathway transcription factor TCF-4 in ileal Crohn's disease.10 However, some of these studies have focused on the responses of monocytes to the inflammatory cytokines in the immune system through Wnt5A or Wnt3A.11, 12 Enteric bacteria and intestinal epithelial cells are known to play key roles in the pathogenesis of IBD.13, 14 The effects and molecular mechanisms of the other Wnt proteins in regulating intestinal inflammation remain unknown.
Previous studies have shown that downstream targets and regulators of the Wnt/β-catenin signaling are involved in bacterial infection. Uropathogenic Escherichia coli attach to the superficial facet cell layer of the bladder epithelium via FimH adhesin. A DNA microarray study found that FimH-mediated attachment suppresses transforming growth factor-beta and Wnt5A signaling to promote subsequent differentiation of basal/intermediate cells.15 Chibby physically interacts with β-catenin to repress its activation of transcription. Chibby knockout mice challenged with Pseudomonas aeruginosa are unable to clear the bacteria from the nasal cavity.16Mycobacterium tuberculosis-infected macrophages had elevated Fz1/Wnt3A expression.17 Infection of gastric epithelial cells with Helicobacter pylori increases the function of the proximal Wnt signaling components low-density lipoprotein receptor-related protein 618 and Dishevelled (Dvl) in the activation of β-catenin.19, 20 Bacteroides fragilis activated the β-catenin pathway in the intestinal epithelia cells.21, 22 We reported that Salmonella activates the Wnt/β-catenin pathway to modulate intestinal inflammation, cell proliferation, and intestinal stem cell niches.5, 6, 23–25
Recent population-based cohort study demonstrates an increased risk of IBD in individuals with Salmonella infection.26 Chronic Salmonella colonization in intestine increases the risk of the intestinal fibrosis, inflammation, and liver abscess in mice.18, 27 However, how bacteria directly regulate individual Wnt family members in intestine and how Wnt modulates the host responses to pathogenic bacteria are unknown.
In the current study we focused on Wnt2 because Wnt2 is known to directly regulate β-catenin activity, the key player in proliferation and inflammation.2–7 We hypothesized that Wnt2 is involved in host protection in intestine by inhibiting inflammation and apoptosis induced by Salmonella infection. Using in vitro and in vivo models, we found that Salmonella-colonized intestinal epithelial cells had elevated mRNA and protein expression levels for Wnt2. Overexpression of Wnt2 is able to lower IL-8 secretion induced by Salmonella infection. Moreover, we identified a bacterial effector protein, AvrA, that is able to regulate Wnt2 expression. Furthermore, the physiological relevance of bacterial Wnt2 activation was investigated in vivo. Immunostaining further identified that activated Wnt2 localized to intestinal epithelial cells. Overall, our studies emphasize the importance of Wnt2 in modulating mucosal inflammation in host–bacterial interactions.
MATERIALS AND METHODS
All animal protocols were approved by the University of Rochester University Committee on Animal Resources and the Institutional Animal Care (No. 2007-065) and Use Committee of the University of North Carolina at Chapel Hill (04-100.0A).
Bacterial Strains and Growth Condition
Bacterial strains used in this study included wildtype Salmonella typhimurium strain ATCC14028 (WT-SL), commensal E. coli F18 (39), nonpathogenic Salmonella mutant strain PhoPc,40 PhoPc AvrA−, PhoPc AvrA−/AvrA+, PhoPc AvrA+, wildtype Salmonella SL1344 (SB300), and its AvrA mutant strain SB1117. E. coli F18AvrA+ was generated by overexpressing the AvrA gene in a pSKW29 low-copy plasmid (pWSK29 was a gift from Dr. Sidney R. Kushner, University of Georgia). Nonagitated microaerophilic bacterial cultures were prepared by inoculating 10 mL of Luria-Bertani (LB) broth with 0.01 mL of a stationary phase culture followed by overnight incubation (≈18 hours) at 37°C, as previously described.5, 41
AvrA Clone and Anti-AvrA Antibody
The avrA gene was from wildtype S. typhimurium strain SL3201. Sequence analysis revealed that the avrA allele used in our study was identical to the allele from S. typhimurium LT2 (GenBank Access. No. AE008830). Anti-AvrA antibody was custom-made as previously described.42
Reverse Transcription Polymerase Chain Reaction (RT-PCR) for AvrA
Total RNA was extracted from bacteria using a Qiagen RNeasy mini kit (Cat. no. 74104; Valencia, CA) according to the manufacturer's protocol. Total RNA was further digested with DNase I (Cat. no. 18068-015; Invitrogen, Carlsbad, CA). RNA integrity was verified by gel electrophoresis. Extracted RNA yield and purity was then determined by measuring absorbances from the 220–350 nm range. RNA reverse transcription was performed using a SuperScript III kit (Invitrogen, Cat. no. 18080-051). cDNA reaction products were then used in a quantitative PCR reaction. The reaction mixture was subjected to PCR amplification using Taq polymerase (Fermentas, Glen Burnie, MD; Cat. no. EP0404) as previously described.43
Human embryonic kidney 293 cells, epithelial Caco2-BBE, HT29Cl.29A, HeLa, and mouse intestinal epithelial CMT93 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS), penicillin-streptomycin, and L-glutamine.24, 38, 42, 44 Human colonic epithelial HCT116 cells were cultured in McCoy's 5A medium supplemented with 10% (v/v) fetal bovine serum (FBS).24, 41 The rat small intestinal IEC-18 cell line was grown in DMEM (high glucose, 4.5 g/L) containing 5% (v/v) FBS, 0.1 U/mL insulin, 50 μg/mL streptomycin, and 50 U/mL penicillin. MEFs were isolated from E13.5 embryos C57BL/6 mouse breeding.24 Briefly, the embryos were harvested and placed in phosphate-buffered saline (PBS) to remove the internal organs, head, and four limbs. The remaining embryo body was individually minced and digested with 0.5% trypsin and 10 mM EDTA for 0.5 hours at 37°C. The digested materials were gently pipetted to a single-cell suspension. The cells were cultured in DMEM (high glucose, 4.5 g/L) containing 10% FBS, 50 μg/mL streptomycin, and 50 U/mL penicillin.
Streptomycin Pretreated Mouse Model
Animal experiments were performed by using specific pathogen-free female C57BL/6 mice (Taconic, Hudson, NY) that were 6–7 weeks old, as previously described.6 The protocol was approved by the University of Rochester University Committee on Animal Resources (UCAR). Water and food were withdrawn 4 hours before oral gavage with 7.5 mg/mouse of streptomycin (75 μL of sterile solution or 75 μL of sterile water in control). Afterward, animals were supplied with water and food ad libitum. Twenty hours after streptomycin treatment, water and food were withdrawn again for 4 hours before the mice were infected with 1 × 107 CFU of S. typhimurium (100 μL suspension in Hank's balanced salt solution [HBSS]) or treated with sterile HBSS (control) by oral gavage as previously described.45, 46 At the indicated times postinfection, mice were sacrificed and tissue samples from the intestinal tracts were removed for analysis.
Germ-free (GF) and Mono-associated Mouse Colon
GF 129/SvEv mice and IL10−/− mice were obtained from the Center for Gastrointestinal Biology and Disease Gnotobiotic Core Facility and the National Gnotobiotic Rodent Resource Center (University of North Carolina, Chapel Hill). Mice were analyzed at the UNC National Gnotobiotic Rodent Resource Center. Sterility of germ-free mice was documented on a monthly basis by fecal Gram stains and aerobic and anaerobic cultures of the feces and bedding. For selected mice, sterility of cecal contents was documented Gram stain and cultures at the time of necropsy. Mono-associated mice were originally GF mice that were infected with a human commensal enteric bacterium E. coli F1839 or E. coli F18AvrA+ by oral swabbing.47
Mouse Colonic Epithelial Cells
Mouse colonic epithelial cells were collected by scraping from mouse colon including proximal and distal regions.6 Cells were sonicated in lysis buffer (1% Triton X-100, 150 mM NaCl, 10 mM Tris [pH 7.4], 1 mM EDTA, 1 mM EGTA [pH 8.0], 0.2 mM sodium ortho-vanadate, protease inhibitor cocktail). The protein concentration was measured using BioRad Reagent (Hercules, CA).
Construction of the Plasmid pcDNA3-mWnt2
The pCMV-Myc-hWnt2 (4.8 kb) plasmid was constructed by inserting a SalI-KpnI fragment containing the human Wnt2 complementary DNA (cDNA) (1.08 kb) into the pCMV-Myc plasmid (3.8 kb). PCR primers for generating a SalI-KapI fragment containing the hWnt2 cDNA were as follows: Sal-hWnt2 forward primer: 5-CGGTCGACATGAA CGCCCCTCTCGGT-3; hWnt2-KpnI backward primer: 5-GGGGTACCTCATGTAGCGGTTGTCCAGT-3.
Transient transfections were performed with Lipofectamine 2000 (Invitrogen) in accordance with the manufacturer's instructions. Briefly, 6 × 105 HCT116 cells were seeded on 60-mm dishes overnight before cotransfection with 4 μg Myc tagged Wnt2. DNA was mixed with the liposome reagent at a ratio of 1:1 before adding to cells. After a 24-hour transfection, proteins were extracted with RIPA buffer for immunoblotting.48
Wnt2 siRNA and control siRNA were purchased from Santa Cruz Biotechnology (Wnt2 siRNA: Cat. no. sc-36841; control siRNA: Cat. no. sc-37007; Santa Cruz, CA). The control siRNA consisted of a scrambled sequence that would not lead to the specific degradation of any cellular mRNA. Caco2-BBE cells were transfected using the DharmaFect 1 transfection reagent (Dharmacon) in Opti-MEM I medium (Invitrogen), according to manufacturer protocols, using a final siRNA concentrations of 50 or 100 nM. After siRNA transfection, the plates were incubated for 3–5 days before further analysis.
Mouse epithelial cells were lysed in lysis buffer (1% Triton X-100, 150 mM NaCl, 10 mM Tris [pH 7.4], 1 mM EDTA, 1 mM EGTA [pH 8.0], 0.2 mM sodium ortho-vanadate, protease inhibitor cocktail). The protein concentration was then measured. MEF cells were rinsed twice in ice-cold HBSS, lysed in protein loading buffer (50 mM Tris (pH 6.8), 100 mM dithiothreitol, 2% SDS, 0.1%bromophenol blue, 10% glycerol), and sonicated. Equal amounts of protein were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to nitrocellulose, and immunoblotted with primary antibodies. The following antibodies were used: anti-c-Myc, anti-PARP (Cell Signal, Beverly, MA), anti-Villin, goat anti-Wnt2 (Santa Cruz Biotechnology), anti-ubiquitin (Biomol, Plymouth Meeting, PA), and anti-β-actin (Sigma-Aldrich, Milwaukee, WI). They were then visualized by ECL. Membranes probed with more than one antibody were stripped before reprobing.
Colonic tissues from the proximal and distal portion of the colon were freshly isolated and embedded in paraffin wax after fixation with 10% neutral buffered formalin. Immunostaining was performed on paraffin-embedded sections (1 μm) of mouse colons. After preparation of the slides as described previously,25 slides were incubated in 3% hydrogen peroxide for 20 minutes at room temperature to block endogenous peroxidase activity and incubated for 20 minutes in 5% BSA with 0.1% saponin in PBS to reduce nonspecific background. Cultured cells were rinsed three times in PBS, fixed for 20 minutes in cold 100% ethanol, and then permeabilized for 10 minutes with 0.2% Triton X-100. Finally, the cells were rinsed three times with PBS containing 10% BSA. The permeabilized cells or tissue samples were incubated with primary antibodies for 90 minutes at room temperature. Samples were then incubated with Dork anti-Goat Alexa Fluor 594 and DAPI (Molecular Probes, Eugene, OR, Invitrogen) for 1 hour at room temperature. Tissues were mounted with SlowFade (SlowFade AntiFade Kit, Molecular Probes) followed by coverslipping. The edges were sealed to prevent drying. Specimens were examined with a Leica SP5 Laser Scanning confocal microscope.
After 24 hours of treatment with Salmonella, 1 × 106 adherent cells were trypsinized and incubated with FITC-conjugated Annexin V (binds to phosphatidyl serine on the cytoplasmic surface of the cell membrane) and propidium iodide (PI) for 15 minutes in the dark according to the manufacturer's protocol (Annexin VFITC Apoptosis Detection kit; Oncogene Research Products, San Diego, CA). Cells were analyzed by flow cytometry.
Salmonella typhimurium Invasion of Human Epithelial Monolayers
Infection of HCT116 cells was performed by a previously described method.45 Bacterial solutions (≈20 bacteria/epithelial cell) were added and bacterial invasion was assessed after 1 hour. Cell-associated bacteria represent bacteria adhered to and/or internalized into the monolayers. These bacteria were released by incubation with 100 μL of 1% Triton X-100 (Sigma). Bacteria internalized in epithelial cells were released with 1% Triton X-100 20 minutes after the addition of gentamicin (50 μg/mL). Gentamicin, an aminoglycoside antibiotic, does not permeate eukaryotic plasma membranes and is, therefore, cytolytic only to extracellular populations of bacteria; intracellular bacterial populations remain viable.49 For both cell-associated and internalized bacteria, 0.9 mL LB broth was then added and each sample was vigorously mixed and quantitated by plating for CFUs on MacConkey agar medium.41
Salmonella-induced Human IL-8 Secretion
HCT116 cells were cultured in DMEM, followed by Salmonella-containing HBSS (1.6 × 1010 bacteria/mL) for 30 minutes, washed three times in HBSS, and incubated at 37°C for 6 hours. Cell supernatants were removed and assayed for IL-8 by enzyme-linked immunosorbent assay (ELISA) in 96-well plates as described previously.45
Real-time Quantitative PCR Analysis
Total RNA was extracted from epithelial cell monolayers using TRIzol reagent (Invitrogen). RNA integrity was verified by gel electrophoresis. RNA reverse transcription was performed using the iScript cDNA synthesis kit (Bio-Rad) according to the manufacturer's directions. The RT cDNA reaction products were subjected to quantitative real-time PCR using the MyiQ single-color real-time PCR detection system (Bio-Rad) and iQ SYBR Green Supermix (Bio-Rad) according to the manufacturer's directions. IL-8 cDNA was amplified using primers to the human IL-8 gene that were complementary to regions in exon 1(5′-TGCATAAAGACATACTC CAAACCT) and overlapping the splice site between exons 3 and 4 (5′-AATTCTCAGCCCTCTTCAAAAA). All expression levels were normalized to actin levels or GAPDH of the same sample using forward (5′-AGA GCA AGA GAG GCA TCC TC-3′) and reverse (5′-CTC AAA CAT GAT CTG GGT CA-3′) primers for actin and the forward (5′-GAC GGC CGC ATC TTC TTG T-3′) and reverse (5′-CAC ACC GAC CTT CAC CAT TTT-3′) primers for GAPDH. Percent expression was calculated as the ratio of the normalized value of each sample to that of the corresponding untreated control cells. All real-time PCR reactions were performed in triplicate. All Wnts and Fzs PCR primers are shown in Table 1.50, 51
|Gene||Primer Sequence||Gene||Primer Sequence|
Data are expressed as mean ± SD. All statistical tests were 2-sided. P-values less than 0.05 were considered statistically significant. Differences between two samples were analyzed by Student's t-test. Statistical analyses were performed using SAS v. 9.2 (SAS Institute, Cary, NC).
Salmonella Increase the mRNA Expression of Wnt2 and Its Receptors in Host Cells
First, we used real-time PCR to investigate Wnt2 mRNA expression in cells before or after S. typhimurium treatment. Figure 1A shows the mRNA expression of Wnt2 and other Wnt proteins in intestinal epithelial IEC-18 cells colonized with S. typhimurium. We found that Wnt2 expression levels in Salmonella-colonized cells were significantly higher than that in the control cells, whereas Wnt2b, 3a, 7b, and 10b were slightly altered by Salmonella colonization. Tumor necrosis factor alpha (TNF-α) is a proinflammatory cytokine. Therefore, we tested if TNF was able to alter Wnt expression after TNF treatment for only 30 minutes, which is considered the early stage of inflammation.28 Our data indicated that TNF-α did not increase Wnt2 expression (data not shown).
Fz and Wnt interact with other structural components at the cell surface to initiate complex signal transduction cascades that culminate in the transcriptional regulation of gene expression.29 We investigated the expression of Fz receptors for Wnt2 in cells with S. typhimurium colonization. Wnt receptors, including Fz4 and 9 for Wnt2,30, 31 were increased upon bacterial colonization (Fig. 1B). These data indicated that the transcriptional levels of Wnt2 and its receptors were enhanced by Salmonella infection in intestinal epithelial cells.
Salmonella Infection Increases the Protein Expression Level of Wnt2 in Intestinal Epithelial Cells
We then examined Wnt2 protein expression after pathogenic S. typhimurium infection in vitro. Using western blot assays, we found that Wnt2 expression increased after infection with Salmonella SL1344 in human colonic epithelial HCT116 cells (Fig. 2A). We tested this effect in different cell lines. A similar change in Wnt2 expression was also found in mouse intestinal epithelial CMT93 cells (Fig. 2B) and human Caco2-BBE cells (Fig. 2C). Taken together, our in vitro data indicated that Wnt2 protein expression was elevated by pathogenic Salmonella colonization in intestinal epithelial cells.
Wnt2 Is Involved in the Regulation of Inflammatory Pathways in Salmonella–Epithelial Interactions
Because our data suggested that Wnt2 is involved in Salmonella colonization in intestinal epithelial cells, we further hypothesized that Wnt2 signaling contributes to inflammatory responses. We transfected a c-myc-tagged Wnt2 plasmid into human epithelial cells and found enhanced Wnt2 protein levels by using an anti-c-myc antibody (Fig. 3A). However, there was no difference in bacterial invasion between normal cells and cells overexpressing Wnt2 (Fig. 3B). In the overexpressed Wnt2 system, we found that the nuclear factor (NF)-κB activity was lower after Salmonella colonization (Fig. 3C). The mRNA expression of the inflammatory cytokine IL-8 was significantly downregulated in the Wnt2 overexpressed cells after pathogenic Salmonella colonization (Fig. 3D). Furthermore, the interleukin (IL)-8 secreted into the cell culture media was detected by ELISA. There was a significantly suppressed IL-8 level in cells overexpressing Wnt2 (Fig. 3E).
To clarify the effect of Wnt2 on the inflammatory response, we further blocked Wnt2 expression by siRNA (Fig. 3F). We found that IL-8 expression was significantly higher in cells with low Wnt2 expression compared to control cells after Salmonella colonization. Hence, these data indicated that Wnt2 expression is able to inhibit Salmonella-induced IL-8 in intestinal epithelial cells.
Overexpressed Wnt2 Protects Cells from Salmonella-induced Apoptosis
We postulated that overexpression of Wnt2 is a strategy for host defense. Apoptosis is one of the biological effects induced by bacterial infection.32, 33 We then determined the percentage of apoptotic cells with different Wnt2 expression statuses. As shown here, Salmonella induced 8.9% later apoptosis in HCT116 cells and 7.3% in cells transfected with an empty pCMY-myc plasmid (Fig. 4A, upper right panel of the histogram). In contrast, Wnt2 overexpressing cells had only 4.6% apoptotic/necrotic cells. The difference of late apoptotic cells was significant between the cells with normal Wnt2 and overexpressed Wnt2 (Fig. 4B). Salmonella induced 58.9% apoptosis/necrosis in HCT116 cells and 59.7% in cells transfected with an empty pCMY-myc plasmid (Figs. 4C, left and right panels of the histogram). In contrast, Wnt2 overexpressing cells had only 25.4% apoptotic/necrotic cells (Fig. 4B). Overall, there was a significant decrease of apoptotic/necrotic cells in the Wnt2 overexpression group (Fig. 4C). The lower right panel in Figure 4A represents early apoptotic cells. We did not find any changes in early apoptotic cells with overexpressed Wnt2. These data indicated that Wnt2 overexpression is able to protect cells from apoptosis and necrosis at late stages.
Wnt2 Protein Degradation Is Regulated by Ubiquitination
Wnt2 is known to be associated with β-catenin activity and cell proliferation.34, 35 We used Wnt2 siRNA to block Wnt2 expression and further tested its effects on the β-catenin pathway in intestinal epithelial cells. We found that β-catenin levels were also low in cells with low levels of Wnt2. Moreover, PCNA, a marker for proliferative cells, was decreased in cells with low Wnt2 levels (Fig. 5A).
We further hypothesized that Wnt2 stabilization is regulated through ubiquitination. To test our hypothesis, we treated cells with MG262, a proteasome inhibitor. We found enhanced Wnt2 ubiquitination in cells after MG262 treatment (Fig. 5B). This result indicated that the stabilization of Wnt2 is regulated through ubiquitination/proteasome. We tested the protein level of Wnt11, the other Wnt family member, with or without MG262 treatment. However, Wnt 11 expression was not changed by MG262 treatment (data not shown).
Salmonella Effector AvrA Upregulates Wnt2 Expression
AvrA is a known bacterial effector protein that inhibits inflammation in bacterial-induced intestinal inflammation by stabilizing β-catenin, the downstream target of Wnt signaling.6, 25 Furthermore, we hypothesized that AvrA is involved in Wnt2 signaling during bacterial colonization. We examined the effects of AvrA on induction of Wnt2 in a human epithelial cell model. PhoPc is a mutant strain derived from Salmonella ATCC14028 and has sufficient AvrA protein expression.25 We used a parental PhoPc AvrA mutant (AvrA−) or the AvrA− complemented strain (PhoPc AvrA−/AvrA+) from previous studies.36 Parental PhoPc strain colonization stabilized Wnt2, whereas cells treated with PhoPc AvrA− bacteria displayed reduced levels of Wnt2. In cells colonized by PhoPc AvrA−/AvrA+, complementary AvrA expression increased Wnt2 protein expression (Fig. 6A). The statistical analysis of Wnt2 intensity further indicated that AvrA expression in host cells significantly stabilized Wnt responses in vitro (Fig. 6A).
Bacterial Colonization Increases Wnt2 in Mouse Intestine In Vivo
To assess the physiological relevance of Wnt2 in Salmonella-induced infection, we investigated the signaling pathway using a Salmonella-colitis mouse model.37, 38 Wildtype strain SL1344 expresses AvrA protein, whereas SB1117 is an AvrA− mutant derived from SL1344. We found that Wnt2 protein expression was upregulated by AvrA expression in SL1344 intestine, whereas there was less Wnt2 in intestine infected with the SB1117 AvrA-mutated bacterial strain (Fig. 6B). Overall, Wnt2 protein expression was significantly enhanced in Salmonella SL1344 infected intestine.
We also investigated the distribution of Wnt2 in mouse intestine before and after Salmonella infection (Fig. 6C). We found clear positive staining for Wnt2 in epithelial cells lining the crypt region as well as the villus and surface epithelium. Pathogenic Salmonella infection enhanced Wnt2 staining. Hence, Salmonella infection not only increased Wnt2 expression but also induced relocation of Wnt2 in intestine. Wnt2 activation is associated with intestinal proliferation. Moreover, we found that PCNA staining, a readout for cell proliferation, was enhanced in the PhoPc and the AvrA-complemented strain PhoPc AvrA−/AvrA+ infected intestine (Fig. 6D).
Bacterial Protein AvrA Enhances Wnt2 Expression
Recently, we found that the AvrA gene existed in a human-isolated commensal E. coli F18 strain (Fig. 7A). The full length of AvrA was amplified in E. coli F18 but not in E. coli JM109. We also identified AvrA mRNA in F18 and various Salmonella strains but not in the Salmonella AvrA− strain (Fig. 7B). AvrA protein expression was found in E. coli strain F18 by western blot analysis using an anti-AvrA antibody (Fig. 7C).
We investigated the role of E. coli AvrA on Wnt expression in germ-free mice and mice mono-associated with E. coli F18 or E. coli F18AvrA+ (overexpressing AvrA). We noted that Wnt2 expression was undetectable in the germ-free intestine, upregulated by E. coli F18 colonization, and further enhanced by E. coli F18AvrA+ mono-associated in the ex-germ-free mouse intestine (Fig. 7D). Moreover, we compared germ-free mice and E. coli F18 mono-associated mice to explore the effect of E. coli on modulating Wnt2 distribution. In E. coli F18 mono-associated small intestine, Wnt2 was not only located at the bottom of the crypts but also in the middle of the villus (Fig. 7E). However, Wnt2 staining was very weak in the germ-free mouse intestine. Taken together, our data indicated that enteric bacteria modulate host responses through the Wnt pathway.
Our current study attempted to answer two questions: is Wnt2 involved in Salmonella infection, and why is Wnt2 signaling involved in host cells during bacterial-induced apoptosis. We have demonstrated that Salmonella infections induce the elevated expression of Wnt2 mRNA and protein. Overexpressed Wnt2 is able to lower the IL-8 induced by Salmonella infection. In contrast, blocking Wnt2 protein expression in cells led to enhanced IL-8 mRNA levels after Salmonella colonization. Compared to normal cells, we found less Salmonella-induced apoptosis in cells with Wnt2 overexpression. Interestingly, we found that Wnt2 protein degradation is associated with ubiquitination. AvrA is a bacterial protein existing in both E. coli and Salmonella. Our study further demonstrated AvrA stabilized Wnt2 in vitro and in vivo. Pathogenic bacteria regulate Wnt2 expression and location in vivo. In an ex-germ-free system, the expression of Wnt2 was dramatically increased by E. coli F18 expressing AvrA. Hence, we have demonstrated that Wnt2 is directly involved in the regulation of intestinal inflammation induced by enteric bacteria.
Our previous studies have demonstrated that β-catenin is a negative regulator of intestinal NF-κB activity in bacterial-induced epithelial inflammation.5, 24 In the canonical pathway, Wnt binding stabilizes the transcription factor β-catenin, which in turn enters the nucleus to regulate Wnt pathway target genes. Our data indicate that Wnt2 may contribute to host protection in intestine by affecting signaling pathways related to cell proliferation and apoptosis by modulating inflammatory responses. Wnt2 signaling leads to downregulation of the JNK/AP-1 and NF-κB pathways, upregulation of β-catenin, and promotes viability in intestinal epithelial cells during bacterial infection. Wnt family members have different functions, even if they sometimes complement each other's functions. We found Wnt2 had no effect on inhibiting bacterial invasion, whereas Wnt11 significantly decreased Salmonella invasion in epithelial cells (data to be published). Taken together, it is now clear that Wnt, its regulators, and downstream target β-catenin are all involved in bacterial inflammation in intestine using different strategies.
Our study showed that the Wnt2 protein is regulated by ubiquitination. Proteasome inhibitor treatment is able to stabilize Wnt2. AvrA has been reported as a deubiquitinase.25 Therefore, AvrA may stabilize Wnt2 by removing the ubiquitin from Wnt2. However, in the AvrA− SB1117 infected intestine, Wnt2 protein expression was still higher than the control intestine without Salmonella infection. This result indicates that other bacterial factors also contribute to upregulation of the Wnt2 protein.
The limitation of our study is that we could only narrow our focus to Wnt2. There are complicated interactions among the 19 Wnt members in this signaling pathway. The other Wnt proteins may synergistically play critical roles in intestinal inflammation. We do not have Wnt2 knockout mice to further investigate Salmonella infections in vivo.
In summary, the current study reports several novel aspects of Wnt2 in host–bacterial interactions: 1) the elevation of Wnt2 by bacterial infection at both the mRNA and protein levels, 2) the protective role of Wnt2 in inhibiting apoptosis and intestinal inflammation during bacterial infection, and 3) the stabilization of Wnt2 by the bacterial protein AvrA from certain Salmonella and E. coli strains. Hence, Wnt2 contributes to host protection in response to pathogenic bacteria in intestine. In addition to providing insights into salmonellosis, our study on the role of Wnt2 in host–bacterial interaction models provides fundamental molecular insights into mucosal inflammation.
We thank Maureen Bower and Bo Liu for technical assistance with gnotobiotic mice and Drs. Nayden Naydenov and Victor Morales for help with apoptosis assays.