Hyperosmolar environment and intestinal epithelial cells: impact on mitochondrial oxygen consumption, proliferation, and barrier function in vitro

Abstract

The aim of the present study was to elucidate the in vitro short-term (2-h) and longer-term (24-h) effects of hyperosmolar media (500 and 680 mOsm/L) on intestinal epithelial cells using the human colonocyte Caco-2 cell line model. We found that a hyperosmolar environment slowed down cell proliferation compared to normal osmolarity (336 mOsm/L) without inducing cell detachment or necrosis. This was associated with a transient reduction of cell mitochondrial oxygen consumption, increase in proton leak, and decrease in intracellular ATP content. The barrier function of Caco-2 monolayers was also transiently affected since increased paracellular apical-to-basal permeability and modified electrolyte permeability were measured, allowing partial equilibration of the trans-epithelial osmotic difference. In addition, hyperosmotic stress induced secretion of the pro-inflammatory cytokine IL-8. By measuring expression of genes involved in energy metabolism, tight junction forming, electrolyte permeability and intracellular signaling, different response patterns to hyperosmotic stress occurred depending on its intensity and duration. These data highlight the potential impact of increased luminal osmolarity on the intestinal epithelium renewal and barrier function and point out some cellular adaptive capacities towards luminal hyperosmolar environment.

Figure 1

Effects of apical hyperosmolarity on cell viability. Assays were performed after treatment during 2- or 24-h with hyperosmotic media or isosmotic control medium. (A) Number of adhering cells measured by cell counting. (B) Cell viability estimated by measuring LDH released in the medium before and after 1% triton treatment and expressed as ratio of alive and total cells. (C) Cell viability evaluated with MTT test, and (D) ADP/ATP cell ratio from the amount of ATP and ADP cell content (expressed as relative luminescence units in the curly bracket). In (A) and (B) experiments, cells were plated on 24-well plates and grown for 3 days before hyperosmotic stress assay. In (C) and (D) experiments, cells were plated on 96-well plates and grown 3 days before hyperosmotic stress assay. Values are from three to four independent experiments (n = 8–32 for each experimental group). Mean significant differences (P < 0.05) are indicated by a different letter. Main factor and interaction effects are indicated with *P < 0.05, **P < 0.01 and ***P < 0.001. NS: Non-significant difference.

Figure 2

Effects of apical hyperosmolarity on trans-epithelial electrical resistance, cell morphology and epithelial permeability to macromolecules markers. Differentiated Caco-2 cells cultured on Tranwell filters for 15 days were treated during 2- or 24-h with hyperosmotic or control isosmotic media applied at the cellular apical side. (A) TER was measured before the hyperosmotic stress (t₀) and after 2- or 24-h treatment. (B) Light microscope images of cell monolayers were acquired in the different experimental conditions using a 20x objective and the default exposition parameters. (C) Paracellular and (D) transcellular permeabilities were estimated by measuring the FD4 and HRP content respectively in the basal side medium at the end of the experiments. Values are from three independent experiments (n = 4–9 for each experimental group). Mean significant differences (P < 0.05) are indicated by a different letter. Main factor and interaction effects are indicated with *P < 0.05, **P < 0.01 and ***P < 0.001. NS: Non-significant difference.

Figure 3

Effects of hyperosmolarity on Caco-2 cells oxygen consumption. Seven days after seeding, cells were cultured in isosmotic or hyperosomotic (50% and 100%) culture media during 2- or 24-h and then isolated and used for basal oxygen (ST3) consumption measurement (A) without any exogenous agent, (B) after addition of the Fo/F1 ATPase inhibitor oligomycin and (C) after addition of the uncoupler FCCP. OXPHOS was calculated for the 2-h hyperosmotic stress and corresponds to the oxygen consumption rate measured in permeabilized cells in presence of 1.5 mM saturated ADP concentration (D). Values are from three independent experiments (n = 5–18 for each experimental group). Mean significant differences (P < 0.05) are indicated by a different letter. Main factor and interaction effects are indicated with *P < 0.05, **P < 0.01 and ***P < 0.001. NS: Non-significant difference.

Figure 4

Delayed effects of apical hyperosmolar media on IL-8 secretion by Caco-2 cells. Secreted IL-8 after 24-h apical hyperosmotic stress was detected by ELISA test on culture media obtained from undifferentiated Caco-2 cells grown for 3 days, or from media recovered at the basal side of differentiated Caco-2 monolayer after 15 days. Values are from three independent experiments (n = 3 for each experimental group). Mean significant differences (P < 0.05) are indicated by a different letter. Main factor and interaction effects are indicated with *P < 0.05, **P < 0.01 and ***P < 0.001. NS: Non-significant difference.

Figure 5

Schematic view of the intestinal epithelial cell response to luminal hyperosmolar environment. Luminal hyperosmolarity is responsible for mitochondrial dysfunctions in colonocytes that are characterized by decreased oxygen consumption and increased proton leak resulting in altered mitochondrial ATP production with a consequent reduction of the ATP intracellular content. This coincides with a slowdown of colonocyte proliferation, alteration of epithelial barrier function and increased release of the pro-inflammatory cytokine IL-8 by colonocytes. The cell shrinkage provoked by increased luminal osmolarity results in an osmoadaptive cell response which is likely to limit the deleterious effects of luminal hyperosmolarity.