Enteroendocrine cells regulate intestinal barrier permeability

Abstract

The intestinal epithelial barrier is essential for nutrient absorption and protection against ingested pathogens and foreign substances. Barrier integrity is maintained by tight junctions, which are sensitive to inflammatory signals, thus creating a feed-forward loop with an increasingly permeable barrier that further drives inflammation and is the hallmark of inflammatory bowel disease. There are currently no therapeutic strategies to improve the intestinal epithelial barrier. We hypothesized that enteroendocrine cells may play an unappreciated role in maintaining barrier integrity. To test this hypothesis, we seeded human intestinal enteroids with genetic loss of enteroendocrine cells on Transwell filters and evaluated transepithelial electrical resistance, paracellular permeability, and the localization and abundance of junctional proteins. We found that enteroendocrine cells were required to maintain a healthy barrier in crypt-like "stem" and villus-like differentiated cultures. In addition, exogenous supplementation of enteroendocrine-deficient cultures with the hormones peptide tyrosine-tyrosine (PYY), and the somatostatin analog octreotide was sufficient to rescue many aspects of this barrier defect both at baseline and in the presence of the inflammatory cytokine tumor necrosis factor. Surprisingly, these improvements in barrier function occurred largely independently of changes in protein abundance of junctional proteins zona occludens 1, occludin, and claudin-2. These findings support a novel role for enteroendocrine cells in augmenting epithelial barrier function in the presence of inflammatory stimuli and present an opportunity for developing therapies to improve the intestinal barrier.NEW & NOTEWORTHY There are no therapies that directly improve the permeability of the intestinal epithelial barrier. This work uses a human intestinal epithelial model system to demonstrate that sensory enteroendocrine cells are necessary for healthy barrier function and that two of their secreted products, peptide YY and somatostatin, are sufficient to improve barrier function at homeostasis and in the presence of inflammatory cytokines. This could provide novel treatments for strengthening the epithelial barrier in human gastrointestinal disease.

Keywords: barrier function; enteroendocrine cells; intestinal organoids; peptide YY; tight junctions.

Figure 1.. EECs regulate barrier function in the intestinal epithelium.

(A) Enteroids were seeded onto semi-permeable Transwell filters to generate 2D cultures. When monolayers were confluent, they were randomly allocated for analysis of stem or differentiated populations. Hormones were added 24 hours prior to barrier function testing. (B, D) TEER and (C, E) permeability to Lucifer Yellow were measured in WT and EEC-deficient enteroid Transwell monolayers in stem (B,C) and differentiated (D,E) conditions. Data reported as mean ± SEM. P-values calculated by one-way ANOVA with Tukey’s multiple comparison’s test. ns, not significant. n = 5–30 wells/group.

Figure 2.. EECs improve TNF-mediated barrier dysfunction.

Stem and differentiated EEC-deficient enteroid monolayers were treated with octreotide (10 μM), and/or PYY (1 μM) concurrently with 150 ng/mL TNF for 24 hours prior to measuring (A, C) TEER and (B, D) permeability to Lucifer Yellow. Data reported as mean ± SEM. P-values calculated by one-way ANOVA with Tukey’s multiple comparison’s test. ns, not significant. n = 5–30 wells/group.

Figure 3.. Tight junction proteins are altered in crypt-like cells upon the loss of EECs.

(A) Immunofluorescence staining and quantification of corrected cell total fluorescence intensity for ZO-1 and Claudin-2 in enteroid monolayers grown in stem conditions with and without treatment with TNF. Scale bars = 10 μm. (B-D) EEC-deficient enteroid monolayers grown in stem conditions were treated with octreotide (10 μM), and/or PYY (1 μM) concurrently with 150 ng/mL TNF for 24 hours prior to quantification of (B) ZO-1 (225 kDa), (C) occludin (55 kDa), and (D) claudin-2 (22 kDa) protein abundance by Western blot. Data reported as mean ± SEM. P-values calculated by unpaired t-test. n = 3 immunofluorescence images per condition; n = 5–6 immunoblots per condition.

Figure 4.. Tight junction proteins are altered in villus-like cells upon the loss of EECs

(A) Immunofluorescence staining and quantification of corrected cell total fluorescence intensity for ZO-1 and occludin in enteroid monolayers after 7 days of differentiation with and without treatment with TNF during the final 24 hours. Scale bars = 10 μm. (B-D) Differentiated EEC-deficient enteroid monolayers were treated with octreotide (10 μM), and/or PYY (1 μM) concurrently with 150 ng/mL TNF for 24 hours prior to quantification of (B) ZO-1 (225 kDa), and (C) occludin (55 kDa) protein abundance by Western blot. Data reported as mean ± SEM. P-values calculated by unpaired t-test. n = 3 immunofluorescence images per condition; n = 3 immunoblots per condition.