Cross-talk between intestinal epithelial cells and immune cells in inflammatory bowel disease

Abstract

Inflammatory bowel disease (IBD) involves functional impairment of intestinal epithelial cells (IECs), concomitant with the infiltration of the lamina propria by inflammatory cells. We explored the reciprocal paracrine and direct interaction between human IECs and macrophages (MΦ) in a co-culture system that mimics some aspects of IBD. We investigated the expression of intercellular junctional proteins in cultured IECs under inflammatory conditions and in tissues from IBD patients. IECs establish functional gap junctions with IECs and MΦ, respectively. Connexin (Cx26) and Cx43 expression in cultured IECs is augmented under inflammatory conditions; while, Cx43-associated junctional complexes partners, E-cadherin, ZO-1, and β-catenin expression is decreased. The expression of Cx26 and Cx43 in IBD tissues is redistributed to the basal membrane of IEC, which is associated with decrease in junctional complex proteins' expression, collagen type IV expression and infiltration of MΦ. These data support the notion that the combination of paracrine and hetero-cellular communication between IECs and MΦs may regulate epithelial cell function through the establishment of junctional complexes between inflammatory cells and IECs, which ultimately contribute to the dys-regulation of intestinal epithelial barrier.

Figure 1. Expression and functionality of connexins in IECs.

(A) Histogram represents the normalized expression of Cxs to GAPDH assessed by qPCR. (B) Densitometry analysis of western blots of Cx expression normalized to GAPDH in IECs. The results are presented as the means ± S.E. of four independent experiments. (C) Representative western blot of Cx26 and Cx43 expression in IECs. (D) Cellular localization of Cxs, scale bar 10 μm. (E) Homo-cellular gap junctional intercellular communication established between IECs, a representative flow cytometry graph where the shift in MFI is shown following co-culture of unlabeled IECs with calcein-labeled IECs; inhibition of this communication is achieved using 100 μM of 18β-GA. (F) Fluorescence Recovery After Photo bleaching of HT-29 cells. The white arrow indicates the target cell in untreated cells (upper panel) and in treated cells with 100 μM of 18β-GA (lower panel), scale bar 5 μm. (G) Histogram analysis of fluorescent intensity of target cell relative to reference cells of untreated (upper panel) and treated cells (lower panel). Error bars represent the fluorescent intensity of each cell based on several measurements calculated by Zeiss software.

Figure 2. Expression and functionality of connexins in monocytes/macrophages THP-1 cell lines.

(A) Histogram represents the expression of connexins in THP-1 cells normalized to GAPDH assessed by qPCR. (B) Representative western blot of Cx26 and Cx43 expression in IECs. (C) Homo-cellular gap junctional intercellular communication established between THP-1 cells. (D) Expression of TLR2 and TLR4 assessed by qPCR. Histogram represents the normalized expression to GAPDH. (E) Densitometry analysis of western blots of NF-κBp65 and COX-2 expression normalized to actin in THP-1 cells. (F) Representative images of western blot of NF-κBp65 and COX-2, respectively. Expression of (G) MMP-2 and MMP-9, (H) IL-1β, (I) TNF-α, as assessed by qPCR. Histogram represents the normalized expression to GAPDH. The results are presented as the means ± S.E. of three independent experiments ((D), for TLR2 p = 0.0057; E, for NF-κB p65 p = 0.04, for COX-2 p = 0.008; (G) for MMP-9 p = 0.063; (H) for IL-1β p = 0.053; (I) for TNF-α p = 0.03). (K) IL-1β and TNF-α protein level determination by ELISA which was performed per manufacturers’ kit protocols. The results are presented as the means ± S.E. of three independent experiments (for IL-1β, p = 0.012 and for TNF-α, p = 0.015).

Figure 3. Hetero-cellular gap junction intercellular communication (GJIC) established between IECs and THP-1 cells.

(A) Hetero-cellular gap junctional intercellular communication established between IECs and THP-1 cells. (B) Histogram represents the flow analysis of three independent experiments. The results are presented as the means ± S.E. of three independent experiments (One-way ANOVA with Tukey correction for multiple comparisons, p < 0.0001).

Figure 4. Connexin and CD68 expression in colon tissues.

The arrows indicate the expression of Cx26 and Cx43 in epithelial cells in normal (upper panels), and in IBD tissues (lower panel). The arrowheads indicate macrophages. Negative control tissues (with exclusion of the primary antibody) is included, scale bar 10 μm. Images captured at a 63x/1.46 Oil Plan-Apochromatic objective. Increase of CD68 expression in infiltrating macrophages in IBD tissues, indicated by arrowheads, is observed as compared to normal colons. Images captured at a 100x Oil objective.

Figure 5. Cellular localization and expression of Cxs and Cxs-Dendra2 in IECs.

(A,C) Expression of Cx26 (43) and Cx26 (43)-Dendra2 in Caco-2 cells and HT-29 cells, respectively. The arrowheads indicate the expression of Cx26 (43)-Dendra2 Scale bar 10 μm. Images captured at a 63x/1.46 Oil Plan-Apochromatic objective. (B,D). Densitometry analysis of western blots of Cxs-Dendra expression normalized to GAPDH in Caco-2 and HT-29 transduced cells, respectively. The results are presented as the means ± S.E. of three independent experiments ((B), for Cx43D p = 0.00015; (D), for Cx26D p = 0.016 and for Cx43D p = 0.02).

Figure 6. Expression of junctional complexes.

(A) Co-localization images for E-cadherin, ZO-1, and β-catenin in Cx43 overexpressing Caco-2 cells. Scale bar 10 μm. Images captured at a 63x/1.46 Oil Plan-Apochromatic objective. (B) Expression of E-cadherin and ZO-1 in colon tissues. Scale bar 10 μm. Images captured at a 63x/1.46 Oil Plan-Apochromatic objective. (C) Protein expression of ZO-1 and E-cadherin in Cx43 overexpressing Caco-2 cells. α-tubulin used as internal control.

Figure 7. Proposed model for regulation of connexin in IBD.