Mechanism of interferon-gamma-induced increase in T84 intestinal epithelial tight junction

Abstract

Interferon-gamma (IFN-gamma) is an important proinflammatory cytokine that plays a central role in the intestinal inflammatory process of inflammatory bowel disease. IFN-gamma induced disturbance of the intestinal epithelial tight junction (TJ) barrier has been postulated to be an important mechanism contributing to intestinal inflammation. The intracellular mechanisms that mediate the IFN-gamma induced increase in intestinal TJ permeability remain unclear. The aim of this study was to examine the role of the phosphatidylinositol 3-kinase (PI3-K) pathway in the regulation of the IFN-gamma induced increase in intestinal TJ permeability using the T84 intestinal epithelial cell line. IFN-gamma caused an increase in T84 intestinal epithelial TJ permeability and depletion of TJ protein, occludin. The IFN-gamma induced increase in TJ permeability and alteration in occludin protein was associated with rapid activation of PI3-K; and inhibition of PI3-K activation prevented the IFN-gamma induced effects. IFN-gamma also caused a delayed but more prolonged activation of nuclear factor-kappaB (NF-kappaB); inhibition of NF-kappaB also prevented the increase in T84 TJ permeability and alteration in occludin expression. The IFN-gamma induced activation of NF-kappaB was mediated by a cross-talk with PI3-K pathway. In conclusion, the IFN-gamma induced increase in T84 TJ permeability and alteration in occludin protein expression were mediated by the PI3-K pathway. These results show for the first time that the IFN-gamma modulation of TJ protein and TJ barrier function is regulated by a cross-talk between PI3-K and NF-kappaB pathways.

FIG. 1.

Effect of IFN-γ on T84 trans-epithelial electrical resistance (TER). (A) Membrane specificity of IFN-γ effect on T84 epithelial resistance. IFN-γ (10 ng/mL) was added to either apical (AP), basolateral (BL), or combined apical and basolateral compartments (AP+BL). Addition of IFN-γ to the basolateral or combined basolateral and apical compartments produced a significant drop in T84 epithelial resistance (n = 3). *p < 0.01 versus control. (B) IFN-γ (10 ng/nl) caused a time-dependent decrease in TER. Data represent means ± SE of epithelial resistance (n = 4). (C) IFN-γ produced a concentration (0, 1, 10, 100 ng/mL) dependent decrease in T84 TER over the 48-h experimental period. *p < 0.0001 versus control. All experiments were repeated three to six times to ensure reproducibility. (D) Time-course effect of IFN-γ (10 ng/mL) on ¹⁴C-inulin flux from apical to basolateral compartments (n = 4). p < 0.0001 IFN-γ versus Control at 48 h.

FIG. 2.

Effect of IFN-γ on PI3-kinase (PI3-K) and Akt activation. (A) IFN-γ (10 ng/mL) caused a significant increase in the ratio of phosphorylated PI3 kinase enzyme to total PI3-K enzyme as measured by FACE at 5 and 10 min (*p < 0.05 compared to control). (B) IFN-γ (10 ng/mL) caused a significant increase in the ratio of phosphorylated Akt enzyme to total-Akt enzyme as measured by FACE. Addition of LY294002 (20 μM) inhibited Akt phosphorylation to levels below control (*p < 0.05 compared to control, **p < 0.05 compared to interferon-γ). LY treatment alone also caused a decrease in baseline phosphorylated Akt. (C) Filter-grown T84 monolayers were treated with IFN-γ (10 ng/mL) for 5 min with or without LY-294002 (20 μM). PI3-K p85 subunit expression in the soluble and insoluble fractions of T84 cell lysate as determined by Western blot analysis was used as an indirect indication of cytoplasmic-to-membrane translocation. Simultaneously run β-actin blots serve as loading control. IFN-γ treatment (10 μg/mL) caused a rapid increase in PI3-K p85 subunit translocation to the insoluble fraction. The PI3-K inhibitor, LY 294002 (20 μM), prevented the PI3-K translocation to the insoluble fraction.

FIG. 3.

Effect of PI3-K inhibitors on the IFN-γ induced drop in T84 epithelial resistance. LY294002 (20 μM) and Wortmannin (100 μM) significantly prevented the IFN-γ (10 ng/mL) induced drop in T84 epithelial resistance at 48 h (n = 4). *p < 0.001 versus controls. **p < 0.001 versus IFN-γ-treated monolayers.

FIG. 4.

(A) Effect of IFN-γ on NF-κB activation as assessed by NF-κB p65 binding the κB DNA-binding site using ELISA-based DNA-binding assay. IFN-γ (10 μg/mL) caused a time-dependent increase in NF-κB p65 DNA binding, with the maximal binding at 8 h. The increase in IFN-γ induced DNA-binding was inhibited by the addition of excess free “cold” κB-binding site containting oligonucleiotide (WT) to IFN-γ treated samples for 8 h. The addition of excess oligonucleiotide containing a mutated κB-binding site had no effect on DNA binding (MT). (B) The effect of IFN-γ on NF-κB-responsive promoter activation. IFN-γ treatment (10 ng/mL) of transfected T84 monolayers resulted in an increase in NF-κB responsive SEAP reporter activity after 8 h, and the increase in activity persisted for more than 24 h. (C) NF-κB inhibitors prevent the IFN-γ induced increase in NF-κB DNA binding. IFN-γ treatment induces increased p65 DNA-binding that is inhibited by the addition of PDTC (100 μM). (D) Effect of the NF-κB inhibitor, PDTC, on IFN-γ-induced drop in T84 epithelial resistance. PDTC (100 μM) significantly inhibited the IFN-γ (10 ng/mL)-induced drop in T84 epithelial resistance (n = 4). *p < 0.01 versus controls. **p < 0.01 versus IFN-γ-treated monolayers.

FIG. 5.

Cross-talk between PI3-kinase and NF-κB pathways (A) The PI3-K inhibitor LY294002 (20 μM) inhibited the IFN-γ-induced increase in NF-κB-responsive promoter activity. *p < 0.01 versus controls. **p < 0.01 versus IFN-γ-treated monolayers. (B) Western blotting of phospho and total-AKT (60 kDA). IFN-γ (10 ng/mL) and EGF treatment (400 ng/mL) caused significant increases in p-AKT levels as compared to controls. Cotreatment with PDTC (100 μM) did not inhibit the increased p-AKT levels in either case. Reprobing for Total-AKT levels are presented as a loading control.

FIG. 6.

Effect of IFN-γ on occludin protein expression in T84 monolayers. Filter-grown T84 monolayers were treated with vehicle, IFN-γ (10 ng/mL) or IFN-γ and inhibitors for a 48-h experimental period. Protein expression was determined by Western blot analysis. IFN-γ caused a down-regulation of occludin protein expression. IFN-γ induced a downregulation of occluding (68 kDa) protein expression was inhibited by (A) the PI3-K inhibitor LY294002 (20 μM) and (B) NF-κB inhibitor PDTC (100 μM). (C) IFN-γ treatment (10 ng/mL) induced a significant increase in the expression of Claudin-1 (22 kDa). (D) IFN-γ treatment (10 ng/mL) induced a significant decrease in Claudin-2 expression (22 kDa). Simultaneously run β-actin (45 kDa) blots are presented as a loading control for each experiment.