Notch signaling modulates MUC16 biosynthesis in an in vitro model of human corneal and conjunctival epithelial cell differentiation

Abstract

Purpose: Notch proteins are a family of transmembrane receptors that coordinate binary cell fate decisions and differentiation in wet-surfaced epithelia. We sought to determine whether Notch signaling contributes to maintaining mucosal homeostasis by modulating the biosynthesis of cell surface-associated mucins in an in vitro model of human corneal (HCLE) and conjunctival (HCjE) epithelial cell differentiation.

Methods: HCLE and HCjE cells were grown at different stages of differentiation, representing nondifferentiated (preconfluent and confluent) and differentiated (stratified) epithelial cultures. Notch signaling was blocked with the γ-secretase inhibitor dibenzazepine (DBZ). The presence of Notch intracellular domains (Notch1 to Notch3) and mucin protein (MUC1, -4, -16) was evaluated by electrophoresis and Western blot analysis. Mucin gene expression was determined by TaqMan real-time polymerase chain reaction.

Results: Here we demonstrate that Notch3 is highly expressed in undifferentiated and differentiated HCLE and HCjE cells, and that Notch1 and Notch2 biosynthesis is enhanced by induction of differentiation with serum-containing media. Inhibition of Notch signaling with DBZ impaired MUC16 biosynthesis in a concentration-dependent manner in undifferentiated cells at both preconfluent and confluent stages, but not in postmitotic stratified cells. In contrast to protein levels, the amount of MUC16 transcripts were not significantly reduced after DBZ treatment, suggesting that Notch regulates MUC16 posttranscriptionally. Immunoblots of DBZ-treated epithelial cells grown at different stages of differentiation revealed no differences in the levels of MUC1 and MUC4.

Conclusions: These results indicate that MUC16 biosynthesis is posttranscriptionally regulated by Notch signaling at early stages of epithelial cell differentiation, and suggest that Notch activation contributes to maintaining a mucosal phenotype at the ocular surface.

Figure 1.

Notch receptors in human corneal and conjunctival epithelial cell cultures. (A) Representative phase contrast micrographs showing HCLE and HCjE cells at three stages of culture: preconfluent, confluent in the absence of serum, and stratified after 7 days of serum supplementation. (B) By Western blot analyses, the intracellular domain of Notch3, but not Notch1 or Notch2, was present in undifferentiated confluent HCLE cells. The Notch3 antibody bound to a major band in the 80 to 90 kDa region corresponding to the intracellular domain, as previously described. Induction of cell differentiation and stratification resulted in binding of the Notch1 antibody to the 100 kDa intracellular domain. Weak binding of the Notch2 antibody was observed after induction of cell differentiation. (C) The intracellular domains of Notch1 to Notch3 were detected in HCjE cells, primarily after induction of cell differentiation and stratification. Experiments were performed in duplicate. Sample lanes were loaded with 50 μg (Notch1), 100 μg (Notch2), and 20 μg (Notch3) total protein. Arrowheads indicate the position of the Notch transmembrane intracellular domain.

Figure 2.

Effect of γ-secretase inhibition on MUC16 biosynthesis during corneal and conjunctival epithelial cell differentiation. (A) By Western blot analyses, inhibition of Notch signaling with DBZ markedly reduced the levels of MUC16 protein, in a concentration-dependent manner, in preconfluent and confluent HCLE cells. Densitometric analyses revealed that treatment of confluent cells with 25 μM DBZ for 24 and 48 hours decreased MUC16 antibody binding up to 90% and 89%, respectively. (B) Similarly to that of HCLE cells, inhibition of Notch signaling with DBZ markedly reduced MUC16 biosynthesis in preconfluent and confluent HCjE cells. By densitometry, treatment of confluent HCjE cells with 25 μM DBZ for 24 and 48 hours decreased MUC16 antibody binding up to 80% and 82%, respectively. Treating HCLE and HCjE cells with DBZ for 4 hours did not affect MUC16 biosynthesis under any culture conditions. Levels of MUC16 biosynthesis in densitometric analyses are relative to DMSO control at 24 hours. Error bars are SEM values based on 2 separate experiments.

Figure 3.

Effect of γ-secretase inhibition on Notch processing in corneal and conjunctival epithelial cells. Incubation of HCLE and HCjE cells with DBZ for 48 hours markedly reduced the levels of Notch3 intracellular domain in a concentration-dependent manner. (A) Densitometric analyses revealed that 25 μM DBZ reduced Notch3 antibody binding in confluent and stratified HCLE cells by 78% and 72%, respectively. (B) In confluent and stratified HCjE cells, the decrease observed was 52% and 74%, respectively.

Figure 4.

Real-time PCR analysis of MUC16 expression in confluent HCLE and HCjE cells incubated with increasing concentrations of DBZ during 4, 24, and 48 hours. In contrast with the marked decrease of MUC16 found at the protein level (Fig. 2), the level of MUC16 transcripts in DBZ-treated cells was not significantly reduced at any time point compared with DMSO control. Y-axis values were calculated using the 2 ^{− ΔΔC_t} method. Levels of MUC16 transcripts in cells treated with DMSO for 24 hours were used as the calibrator. Error bars are SEM values based on 2 separate experiments.

Figure 5.

Effect of γ-secretase inhibition on MUC1 and MUC4 biosynthesis during corneal and conjunctival epithelial cell differentiation. (A) By Western blot analyses, treatment of HCLE cells with increasing concentrations of DBZ for 4, 24, and 48 hours did not alter the levels of MUC1 and MUC4 compared with DMSO control at any stage of cell differentiation. Due to low levels of MUC4 in human corneal epithelium, amounts of MUC4 protein in HCLE cells were detected using an enhanced chemiluminescent substrate (SuperSignal West Femto Maximum Sensitivity Substrate). (B) Similarly to that of HCLE cells, no change was detected for MUC1 and MUC4 in HCjE cells cultured with DBZ.