Role of epithelial integrin-linked kinase in promoting intestinal inflammation: effects on CCL2, fibronectin and the T cell repertoire

Abstract

Background: The role of integrin signaling in mucosal inflammation is presently unknown. Hence, we aimed to investigate the role of epithelial-derived integrin-linked kinase (ILK), a critical integrin signaling intermediary molecule, in colonic inflammation.

Methods: Conditional intestinal epithelial cell ILK knockout mice were used for assessment of acute and chronic dextran sodium sulfate (DSS)-induced colitis. Disease activity was scored using standard histological scoring, mucosal cytokines were measured using ELISA, chemokines were determined using reverse-transcription polymerase chain reaction, as well as Q-PCR, and intracellular cytokine staining performed using FACS analysis.

Results: In both acute and chronic DSS-induced colitis, compared to wild-type mice, ILK-ko mice exhibit less weight loss, and have reduced inflammatory scores. In an in vitro model system using HCT116 cells, we demonstrate that si-RNA mediated down-regulation of ILK results in a reduction in monocyte chemoattractant protein 1 (MCP1, CCL2) chemokine expression. A reduction in CCL2 levels is also observed in the tissue lysates of chronically inflamed colons from ILK-ko mice. Examination of mesenteric lymph node lymphocytes from ILK-ko mice reveals that there is a reduction in the levels of IFN gamma using intracellular staining, together with an increase in Foxp3+ T regulatory cells. Immunohistochemistry demonstrates that reduced fibronectin expression characterizes the inflammatory lesions within the colons of ILK-ko mice. Intriguingly, we demonstrate that fibronectin is directly capable of downregulating T regulatory cell development.

Conclusions: Collectively, the data indicate for the first time that ILK plays a pro-inflammatory role in intestinal inflammation, through effects on chemokine expression, the extracellular matrix and immune tolerance.

Figures

Figure 1

ILK is induced by pro-inflammatory stimuli. A. Colonic SW480 cells were stimulated with 1 ug/ml LPS for 24 h, and after harvesting the cells, western immunoblotting was performed. Membranes were probed with antibodies for ILK, ser473Akt, Akt and beta-actin. ILK protein induction is accompanied by an increase in the phosphorylation of Akt at ser473, but not the protein level of this kinase. The beta-actin is shown as an internal control. B. This shows the response of a murine colonic explant to exposure with 2.5% DSS for 48 h. There is an increase in the expression of ILK. The lower panel shows equivalent levels of beta-actin. C. Colons of control and DSS treated mice (48 h) were probed with a polyclonal antibody to ILK. Besides an increase in the intensity of the signal generally, more than 50% of the epithelial cell nuclei stain positively for ILK. D and E. Colonic HCT116 cells were stimulated with 5 ng/ml IL-1β for 4 h. Protein was obtained by lysing cells in homogenization buffer and western blotting performed for ILK expression (D). Alternatively, RNA was obtained by Trizol and used for cDNA synthesis; semi-quantitative RT-PCR for ILK was then performed using the primers indicated in materials and methods. Wortmannin and Ly294002 were both capable of inhibiting the IL-1β-induced ILK protein and mRNA. The β-actin signal is shown as the internal control. F. The experiment was repeated as in D with the addition of inhibitors of the MAPK pathway. Specifically PD98059, SP600125 and SB203580 were used to inhibit the p42/44 ERK, P54/45 JNK and p38SAPKs respectively, and GAPDH used as the loading control. Data are representative of 3 independent experiments for each panel.

Figure 2

Epithelial ILK provides a pro-inflammatory stimulus in acute colitis. A. Acute colitis was induced in mice using 3.5% DSS in the drinking water. Mice were weighed daily, and the results after 7 days of DSS treatment, and following a further 8 days off DSS are shown. The weights expressed as the percentage of original (mean ± SEM) are shown (1 = WT control, 2 = WT+DSS, 3 = ILK-ko (C), 4 = ILK-ko + DSS). There is a significant attenuation of the normally observed weight loss in the ILK-ko mice at day 7 (p < 0.05, ANOVA, Tukey post-test). B. Representative histological slides for each of the DSS treated groups are shown. The inflammatory infiltrate, surface epithelial disruption and edema are reduced in the ILK-ko animals. There was a significant reduction in the inflammatory scores in the ILK-ko mice (n = 6 per group).

Figure 3

ILK regulates CCL2 expression. A. Si-RNA was used to knockdown ILK and a scrambled oligonucleotide was used as a control. RNA was extracted from HCT116 cells after 4 h of stimulation with IL-1β and reverse transcribed to make cDNA. This was used to determine the message of the molecules shown using RT-PCR. CCL2 induction in response to the cytokine IL-1β was the only molecule whose expression was blocked by si-RNA knockdown of ILK (see CCL2 panel, lane 4). This result was reproduced on 2 further occasions. B. The experiment was repeated as in A except Q-PCR used to determine the expression of CCL2, either following si-RNA knockdown of ILK (left panel) or using the QLT0267 inhibitor (right panel). C. HCT116 cells were pretreated with the specific ILK inhibitor QLT0267 for 1 hour before being stimulated with IL-1β for the times indicated. CCL2 concentrations were determined in the culture supernatent using ELISA. The data are representative of two separate experiments, each performed in triplicate (*p < 0.01).

Figure 4

ILK regulates wasting disease and CCL2 expression in chronic colitis. A. Chronic DSS (2.5%) colitis was induced in 6 animals per group using 3 rounds of 2.5% DSS in the drinking water. Animal weights were monitored and the data in A show that there was a significant attenuation of the weight loss in the ILK-ko animals (squares), compared with the wild-type group (diamonds), (*p < 0.05). B. Assessment of microscopic damage in these mice indicated a significant reduction in the ILK-ko mice (*p < 0.05). Representative histology is shown in C, where an obvious loss of crypts, mucosal edema and inflammatory infiltrate are shown in the control sample. D. Total RNA was extracted from sections of distal colon and reverse transcribed to make cDNA. This was used to determine the message levels of CCL2 using semi-quantitative RT-PCR. As can be seen from this figure, CCL2 induction was significantly (*p < 0.05) reduced in the ILK-ko animals (the barchart E shows the message levels corrected for β-actin, and are for the entire sets of 6 mice per treatment group). F. Distal colonic lysates were obtained using homogenization buffer and then used to determine CCL2 levels by ELISA. The data are for 6 mice per group and show a reduction in levels of CCL2 in the ILK-ko mice (**p < 0.01).

Figure 5

Fibronectin regulates CCL2, ILK and α5 integrin. HCT116 cells were plated on tissue culture plates coated with increasing levels of fibronectin for 4 h. In A, the CCL2 levels were measured in the medium using ELISA. In B, the cells were subsequently lysed and equivalent amounts of protein resolved using SDS page. After transfer the membranes were probed with the antibodies indicated. The data indicate increased ILK and α5 integrin expression with increasing fibronectin exposure, peaking at 20 ug/ml. C. Immunohistochemistry was performed on colonic sections using a fibronectin antibody on 3 different mice in each group. There is a clear reduction in fibronectin expression in the ILK knockout mice (a representative slide is shown). D. HCT116 cells were exposed to the doses of fibronectin indicated and the QLT 0267 compound added at the start of the experiment. The cell lysates were then resolved using SDS-PAGE and probed with either the α5 integrin or beta actin (loading control) antibodies. The data are representative of 3 separate experiments.

Figure 6

Epithelial ILK regulates tissue expression of inflammatory cytokines. A. Interferon gamma, tumor necrosis factor alpha and interleukin-12p40 cytokine levels were determined in colonic homogenates from 6 ILK-knockout animals and 6 wild-type controls (**p < 0.01). B. Lymphocytes were obtained from mesenteric lymph nodes of wild-type and ILK-ko mice. Intracellular staining for FoxP3 and IFNγ was performed as described in materials and methods. After stimulation with PMA (25 ng/ml) and ionomycin (1 mg/ml) for 6 h, cells were fixed and permeabilized. Then they were stained with the indicated antibodies and read on a BD FACS Canto. The data are from 6 ILK-ko and 6 wild-type mice (*p < 0.05). C. Tissue sections were obtained from control and ILK-ko mice at the end of 3 rounds of DSS treatment, and processed for immunohistochemistry. Using anti-CD3 and anti-FoxP3 antibodies, the number of positively staining cells were counted in 3 fields from 6 separate animals, in each group. The ratios obtained are shown in D (*p < 0.05). E. IL-17A staining was performed using immunofluorescence as described in methods for tissue sections from the same sets of mice as in C. The red staining cells are clearly observed to be more numerous in the control samples, and the data is graphically represented in F.

Figure 7

Fibronectin regulates Treg differentiation. T lymphocytes (CD4+CD25-) were obtained from the mesenteric lymph nodes of mice and exposed to plate bound anti-CD3 (10 mg/ml) and fibronectin, as well as soluble anti-CD28 (1 mg/ml), rIL-2 (100 u/ml) and TGFβ (10 ng/ml) as described in materials and methods. After 24 h the proportion of FoxP3+ cells were determined using FACS analysis. In A, i: representative neutral, 1000 ng/ml fibronectin; ii: iTreg, 0 ng/ml fibronectin; iii: iTreg, 125 ng/ml fibronectin; iv: iTreg, 1000 ng/ml fibronectin. The barchart in B represents the overall dose-response relationship, performed in triplicate and repeated twice. (*p < 0.05, **p < 0.01, using ANOVA).