The CD40-CD40L pathway contributes to the proinflammatory function of intestinal epithelial cells in inflammatory bowel disease

Abstract

In inflammatory bowel diseases (IBD), intestinal epithelial cells (IECs) are involved in the outbalanced immune responses toward luminal antigens. However, the signals responsible for this proinflammatory capacity of IECs in IBD remain unclear. The CD40/CD40L interaction activates various pathways in immune and nonimmune cells related to inflammation and was shown to be critical for the development of IBD. Here we demonstrate CD40 expression within IECs during active IBD. Endoscopically obtained biopsies taken from Crohn's disease (n = 112) and ulcerative colitis patients (n = 67) consistently showed immunofluorescence staining for CD40 in IECs of inflamed ileal or colonic mucosa. In noninvolved mucosa during active disease, tissue obtained during Crohn's disease or ulcerative colitis in remission and biopsies from healthy controls (n = 38) IECs almost entirely lacked CD40 staining. Flow cytometry and RT-PCR analysis using different intestinal epithelial cell lines (HT29, SW480, and T84) showed IFN-gamma to effectively induce CD40 in IECs. Cells were virtually unresponsive to LPS or whole E. coli regarding CD40 expression. In addition, a moderate induction of CD40 was found in response to TNF-alpha, which exerted synergistical effects with IFN-gamma. CD40 ligation by CD40L-transfected murine fibroblasts or soluble CD40L increased the secretion of IL-8 in IFN-gamma pretreated HT29 cells. Our findings provide evidence for the epithelial expression and modulation of CD40 in IBD-affected mucosa and indicate its involvement in the proinflammatory function of IECs.

Figure 1

Expression of CD40 in intestinal epithelial cells during active crohn disease (CD) and ulcerative colitis (UC). Immunofluorescence staining for CD40 (green) was performed on cryosections of endoscopically obtained biopsies. In the healthy ileum (A) and colon (B) IECs lacked staining for CD40. In contrast, CD40 was detected on IECs during CD ileitis (C), CD colitis (D), and UC (E). Epithelial staining for CD40 was found in crypts (asterisks), villi, and the colonic surface epithelium. On the subcellular level CD40 was predominantly localized on basolateral membranes of IECs. Cells of the lamina propria (LP) revealed CD40 staining independent of the inflammatory state. Lu indicates lumen.

Figure 2

Quantification of the CD40 expression in intestinal epithelial cells of IBD patients and healthy controls. The expression of CD40 in IECs was analyzed using immunofluorescence microscopy on endoscopically obtained mucosa of healthy controls and IBD patients. The immunofluorescence staining was determined and quantified as described in the Materials and Methods. CD40 was hardly ever detectable within IECs of the healthy ileum and entirely absent from the colonic epithelium of healthy controls (A). In the majority of patients with active CD ileitis (100%) and colitis (85%), CD40 was consistently found in IECs as opposed to the corresponding patients in remission (ileum 20%; colon 5%; B). Similar results were obtained in UC patients (active disease 92.5%; remission 3.70%; C). In contrast to the inflamed tissue in CD and UC the noninvolved adjacent mucosa almost entirely lacked the epithelial expression of CD40 (D).

Figure 3

Influence of IFN-γ and TNF-α on the surface-expression of CD40 in intestinal epithelial cells. Responsiveness of HT29 (A and B), SW480 (C and D), and T84 (E and F) cells to 72 hours IFN-γ (200 U/ml) and/or TNF-α (1–100 ng/ml) were examined by flow cytometry. Histograms show one representative experiment of each cell line applying 200 U/ml IFN-γ and 100 ng/ml TNF-α (A, C, and E). Diagrams integrate three independent experiments, and results are displayed as mean ± SEM (B, D, and F); Asterisks indicate isotype controls. Compared with cells without any treatment, IFN-γ was highly effective in stimulating CD40, particularly in HT29 and T84 cells (A–F). A dose-dependent moderate response to TNF-α was observed in SW480 and T84 cells (C–E), whereas HT29 cells were unresponsive to this cytokine given alone (A and B). IFN-γ and TNF-α exerted synergistic effects in HT29 and SW480 cells (A–D), which was not seen in T84 cells (E and F). *P ≤ 0.05 compared to untreated cells. **P ≤ 0.05 compared with IFN-γ–treated cells.

Figure 4

CD40 mRNA expression in intestinal epithelial cells after treatment with IFN-γ and TNF-α. Expression levels of CD40 mRNA were assessed in HT29, SW480, and T84 cells using RT-PCR. Results represent the relative gene expression of CD40 and are shown as copies per 100 copies of the metastatic lymph node gene (MLN) 51. Diagrams display means of three independent experiments performed in duplicate. Error bars indicate ± SEM. 200 U/ml IFN-γ caused a marked increase of the CD40 mRNA levels in all cell lines (A–E). In HT29 (A) and SW480 cells (C) highest levels were observed after treatment periods of 72 hours. In T84 cells highest levels were achieved after 24 hours (E). The appropriate periods were adopted for IFN-γ and TNF-α exposure (B, D, and F). Though less effective than IFN-γ, TNF-α (10 and 100 ng/ml) dose-dependently up-regulated the CD40 mRNA in SW480 and T84 (D and F) but not in HT29 cells (B). The combination of IFN-γ and TNF-α was found to be the most effective stimulus. In HT29 and SW480 highest expression levels were seen after treatment with 200 U/ml IFN-γ and 100 ng/ml TNF-α (B and D), whereas in T84 cells the additive influence of 10 ng/ml TNF-α was most effective (F).

Figure 5

Effects of IFN-γ and sCD40L on the CD40 induction in intestinal epithelial cells. Expression levels of CD40 mRNA were determined in HT29 cells and results presented as relation to the metastatic lymph node gene (MLN) 51 mRNA expression. Data are shown as means of three independent experiments performed in duplicate and error bars indicate ± SEM. A: Cells were pulsed with 200 U/ml IFN-γ for 24 hours, and the supernatants were obtained and transferred to naïve cells. In these cells CD40 mRNA expression levels were assessed 4 and 24 hours after transfer without detection of any induction. Cells exposed 72 hours to 200 U/ml IFN-γ served as control and displayed a marked up-regulation. B: In contrast to IFN-γ, sCD40L (100 ng/ml) given for 72 hours did not stimulate CD40 mRNA in naïve cells. Furthermore sCD40L (100 ng/ml given 4, 24, and 72 hours) was ineffective to further increase the CD40 expression subsequent to a IFN-γ pulse for 72 hours.

Figure 6

Effect of IFN-γ and LPS on the surface expression of CD40 in intestinal epithelial cells. The influence of IFN-γ and LPS on the expression of CD40 was assessed in HT29 (A and B), SW480 (C and D), and T84 cells (E and F) using flow cytometry. Cells were exposed to IFN-γ (200 U/ml), LPS (10–1000 ng/ml), or the combination of both stimuli for 72 hours. Histograms display one representative experiment using 200 U/ml IFN-γ and 1000 ng/ml LPS (A, C, and E). Diagrams depict the results of three independent experiments shown as mean ± SEM (B, D, and F). Asterisks indicate isotype controls. In all cell lines CD40 was not detectable without stimulus but strongly induced by IFN-γ (A–E). LPS given alone did not exert inductive effects. Some synergistical effects were observed in SW480 cells when LPS was exposed in parallel with IFN-γ (C and D). *P ≤ 0.05 difference compared with untreated cells.

Figure 7

Bacterial effects on CD40 and IL-8 mRNA expression in intestinal epithelial cells. Expression levels of CD40 (A) and IL-8 mRNA (B) were determined in HT29 cells and results presented as relation to the mRNA expression of the metastatic lymphnode gene (MLN) 51. Data are shown as means of three independent experiments performed in duplicate, and error bars indicate ± SEM. A: Neither LPS (10 to 1000 ng/ml) nor whole E. coli stimulated CD40 mRNA expression after 4 and 48 hours. The response to 200 U/ml IFN-γ served as control. B: In contrast to CD40 mRNA, LPS (10 ng/ml) and E. coli strongly increased the expression levels of IL-8 mRNA after 4 hours.

Figure 8

Mucosal expression of IL-8 in CD and UC. Immunofluorescence staining for IL-8 (green) was performed on cryosections of tissue obtained during ileocoloscopy. Sections made from the healthy ileum (A) or colon (B) lacked mucosal staining for IL-8. Tissue taken during active CD ileitis (C), CD colitis (D), and UC (E) revealed IL-8 staining in the epithelium and within cells of the lamina propria (LP). Lu indicates lumen.

Figure 9

IL-8 secretion by intestinal epithelial cells on CD40 engagement. HT29 cells were pretreated with 200 U/ml IFN-γ to induce the surface expression of CD40. CD40 ligation was performed for 72 hours using either murine fibroblasts (L cells) stably transfected with human CD40L (A) or human recombinant soluble CD40L (B; for details see Materials and Methods). Concentrations of IL-8 were determined in cell-free supernatants by ELISA. IFN-γ dose-dependently enhanced the constitutive secretion of IL-8 (A and B). Compared with nontransfected L cells, the coculture with L-CD40L cells strongly increased the IFN-γ–stimulated IL-8 production (A). The most distinctive increase (170%) was observed in HT29 cells treated with 50 U/ml IFN-γ, the lowest dose applied. Similar to the cell-bound activation of CD40, the exposure of sCD40L was also found to further augment the IFN-γ–stimulated IL-8 secretion up to 44% (B). Results are presented as mean of three independent experiments performed in duplicate. Error bars indicate ± SEM. *P ≤ 0.05 comparing L versus L-CD40L cell coculture (A) and IFN-γ–stimulated versus additionally sCD40L-stimulated cells (B).