Abstract
Abstract: :
Purpose: N–cadherin forms the adherens junction in RPE cells but for unknown reasons E–cadherin localizes to junctions only in some postconfluent cells. Here the subcellular distribution of E– and N–cadherin was analyzed in early cultures when adhesions first form to identify the fate of non–junctional E–cadherin. Methods: Cadherins in hTERT–RPE1 cells at 30 hrs post plating were analyzed by immunostaining for organelle co–distribution and by protein blotting without and with treatment with various inhibitors of protein trafficking and turnover. Results: At 30 hr the 120 kDa processed form of N–cadherin distributes largely to forming junctions with no significant organelle localization. Unprocessed and/or truncated forms of E–cadherin localize not to junctions or surface membranes but rather co–distribute with markers for ER (weakly), lysosomes, proteasomes, autophagosomes, centrosomes and mitochondria but not Golgi, early endosomes or peroxisomes. Inhibitors of actin and N– (but not O–) linked glycosylation disrupt N–cadherin junctions but have no effect on E–cadherin. Inhibitors of lysosomes, furin proprotein convertases, ER–Golgi trafficking, and ER–cytosol Ca++ homeostasis affect both cadherins, variably disrupting N–cadherin junctions and modifying the amount and/or organelle localization of E–cadherin peptides. Proteasome and caspase 3 inhibitors increase immunostaining only of E–cadherin. Inhibitors of microtubules or palmitoylation affect neither cadherin. No treatment results in E–cadherin trafficking to junctions. Conclusions: The results suggest, largely as expected, that a furin–processed, N–glycosylated N–cadherin trafficks efficiently from the ER to junctions in RPE cells. E–cadherin, however, appears to be aberrantly processed and then targeted to various (some unusual) degradative compartments, which may prevent its accumulation in forming junctions and thereby limit its signaling potential in RPE cells.
Keywords: retinal pigment epithelium • cell adhesions/cell junctions