AhR activation defends gut barrier integrity against damage occurring in obesity

Abstract

Objective: Obesity is characterized by systemic and low-grade tissue inflammation. In the intestine, alteration of the intestinal barrier and accumulation of inflammatory cells in the epithelium are important contributors of gut inflammation. Recent studies demonstrated the role of the aryl hydrocarbon receptor (AhR) in the maintenance of immune cells at mucosal barrier sites. A wide range of ligands of external and local origin can activate this receptor. We studied the causal relationship between AhR activation and gut inflammation in obesity.

Methods: Jejunum samples from subjects with normal weight and severe obesity were phenotyped according to T lymphocyte infiltration in the epithelium from lamina propria and assayed for the mRNA level of AhR target genes. The effect of an AhR agonist was studied in mice and Caco-2/TC7 cells. AhR target gene expression, permeability to small molecules and ions, and location of cell-cell junction proteins were recorded under conditions of altered intestinal permeability.

Results: We showed that a low AhR tone correlated with a high inflammatory score in the intestinal epithelium in severe human obesity. Moreover, AhR activation protected junctional complexes in the intestinal epithelium in mice challenged by an oral lipid load. AhR ligands prevented chemically induced damage to barrier integrity and cytokine expression in Caco-2/TC7 cells. The PKC and p38MAPK signaling pathways were involved in this AhR action.

Conclusions: The results of these series of human, mouse, and cell culture experiments demonstrate the protective effect of AhR activation in the intestine targeting particularly tight junctions and cytokine expression. We propose that AhR constitutes a valuable target to protect intestinal functions in metabolic diseases, which can be achieved in the future via food or drug ligands.

Keywords: Aryl hydrocarbon receptor; Cell junction; Intestine; Permeability; Signaling.

Figure 1

Low expression of AhR target genes in obese subjects with intestinal inflammation. (A) CD3⁺ T cell density (cell/mm²) in the epithelium/lamina propria ratio (Epi/LP) was determined by immunohistochemistry in the jejunum of non-obese (n = 10) and obese subjects (n = 26). The results are expressed as mean ± SEM, ∗p < 0.05. (B–F) Pearson's correlations of AhR (B), CYP1A1 (C), IL-22 (D), CYP1B1 (E), AhRR, and (F) mRNA levels in the intestinal epithelium and CD3 Epi/LP ratios in female obese subjects. AhR: aryl hydrocarbon receptor, AhRR: aryl hydrocarbon receptor repressor, a.u.: arbitrary units. Pearson's r and p values corrected for age and BMI of obese subjects are indicated.

Figure 2

Protective role of AhR activation on cell-cell junctions in murine intestinal epithelium. Mice were submitted to 5 gavages (5x) with 200 μl water or palm oil or palm oil + βNF for five consecutive days. (A) The expression of CYP1A1 and CYP1A2 was determined by RT-PCR in the jejunum. Cyclophilin was used as a reference gene. The results are expressed in arbitrary units (a.u.) as the ratio of the target gene to cyclophilin (cyclo) mRNA level as mean ± SEM, n = 5, ∗∗∗p < 0.001 compared to water conditions. (B) The distribution of tight junction proteins ZO-1, occludin, and tricellulin was analyzed by immunofluorescence on the jejunum sections. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). The rectangle indicates the field used for enlargement. The white arrowheads indicate the labeling of the junction proteins. Scale bar = 20 μm. (C) The intestinal permeability was assessed after the last gavage with water, palm oil, or palm oil + βNF by measuring plasma concentrations of 4 kDa FITC-dextran (FD4) 1 h after an oral FD4 load. The results are expressed in μg/ml as mean ± SEM, ∗p < 0.05, and ∗∗p < 0.01 compared to water. Data show one experiment representative of two independent experiments conducted with 4–5 mice in each group.

Figure 3

Protective role of AhR activation on EGTA-induced damage to barrier integrity in Caco-2/TC7 cells. (A) Caco-2/TC7 cells were pre-incubated without (Ctrl) or with 20 μM βNF for 4 consecutive days before the addition or 4.5 mM EGTA for additional 24 h βNF treatment was maintained during all of the experiments. The expression of CYP1A1 was quantified by RT-qPCR. Cyclophilin was used as a reference gene. The results are expressed in arbitrary units (a.u.) as the ratio of the target gene to cyclophilin (cyclo) mRNA level as mean ± SEM, n = 6–15 from 3 independent experiments, ∗∗∗∗p < 0.0001 compared to controls. (B) The cells were cultured under the same conditions as in (A). Paracellular permeability across Caco-2/TC7 cell monolayers was evaluated by measuring the accumulation during 4 h of 4 kDa FITC-dextran (FD4) in the basal compartment. The results are expressed as percentage of FD4 input (amount added in the apical compartment), mean ± SEM, n = 6–15 from 3 independent experiments, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 compared to controls. (C) The cells were cultured under the same conditions as in (A). TEER was assessed in the control and treated cells. The results are expressed in ohm.cm² as mean ± SEM, n = 6–15 from 3 independent experiments, ∗∗p < 0.01, and ∗∗∗p < 0.001 compared to controls. ###p < 0.001 compared to EGTA. (D) The cells were cultured under the same conditions as in (A). Immunofluorescence analysis was conducted to study the location of ZO-1, occludin, and tricellulin. Nuclei were stained DAPI, scale bar = 20 μm (E) The cells were cultured as in (A). Occludin and tricellulin protein levels were determined in cell lysates using capillary-based Western blotting. Reconstituted images are displayed. Hsc70 protein levels were used as loading controls. (F) Caco-2/TC7 cells were first incubated in the presence or absence of 4.5 mM EGTA for 4 h. The media were then removed and replaced by fresh media in the presence or absence of 20 μM βNF for 20 h. TEER was measured before EGTA treatment (T0), after 4 h of EGTA treatment (T4), and at different times after βNF addition. The results are expressed in percentage of TEER measured at T0 for each cell culture condition as mean ± SEM, n = 6 from 2 independent experiments. ∗p < 0.05 and ∗∗∗∗p < 0.0001 compared to controls and ##p < 0.01 compared to EGTA.

Figure 4

AhR activation prevents the increase of cytokine expression induced by EGTA. Caco-2/TC7 cells were cultured under the same conditions as in Figure 3. The mRNA levels of TNF-α (A), IL-1β (B), and IL-8 (C) were quantified by RT-PCR. Cyclophilin was used as a reference gene. The results are expressed in arbitrary units (a.u.) as the ratio of the target gene to cyclophilin (cyclo) mRNA level as mean ± SEM, n = 20–30. (D) The concentration of IL-8 protein in the basal compartment was quantified by ELISA 24 h after EGTA treatment. The results are expressed in pg/ml as mean ± SEM, n = 6. Fold increase compared to the control condition is indicated at the top of the bar plots. ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 compared to controls unless otherwise indicated. Figure 4A,B represent data compiled for 6 independent experiments. Each experiment consisted of 3–6 independent Transwells. Figure 4C represents data compiled for 2 independent experiments conducted in triplicate.

Figure 5

Signaling pathways are involved in the preventive effect of AhR activation on transepithelial resistance. Caco-2/TC7 cells were pre-incubated with (A) 5 μM of protein kinase C inhibitor Ro 31–8220 (Ro), (B) 20 μM of p38MAPK inhibitor SB203580 (SB), or (C) 10 μM of protein kinase ERK1/2 inhibitor U0126 1 h prior βNF addition. These treatments were repeated daily for 4 days. EGTA was added for the last 4 h of treatment. Transepithelial resistance (TEER) was measured at the end of the experiment. The results are expressed in percentage of TEER value measured in control conditions at the end of the experiment as mean ± SEM, n = 3. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 compared to controls unless otherwise indicated. Data are representative of 2 independent experiments conducted in triplicate. (D) Lactate dehydrogenase (LDH) was quantified in the apical medium and the results are expressed as percentage of LDH activity measured in the total cell lysate (mean ± SEM), n = 6.

Figure 6

Protective role of AhR on intestinal barrier and cytokine expression. (A) Obese non-diabetic subjects were classified according to their small intestinal inflammation score established by the densities of T lymphocytes using the CD3 marker. The epithelium/lamina propria (Epi/Lp) CD3 ratio quantified T lymphocyte infiltration within the epithelium in the jejunum samples. Obese subjects with high CD3 Epi/Lp ratios displayed lower AhR activity. (B) In the intestinal epithelial cells, lipid load or treatment with calcium chelator (EGTA) damaged barrier integrity (increase in paracellular permeability and alteration of tight junctions) and induced the expression and secretion of pro-inflammatory cytokine IL-8. In the human intestinal epithelial Caco-2/TC7 cell line, the activation of AhR by an agonist increased the expression of occludin (occ) and tricellulin (tric) and improved their localization at tight junctions (TJ), restored paracellular permeability to ions, and prevented the IL-8 expression and secretion induced by barrier disruptor agent through mechanisms involving protein kinases C and p38MAPK.