Figure 1.
Addition of the PIKfyve inhibitor YM201636 results in intracellular accumulation of claudin-1.
YM201636 treated MDCK cells showed no difference in localisation of ZO-1, occludin, E-cadherin or PKCι/ζ when compared with vehicle treated control cells (A–B left panels and C–E). In contrast, the localisation of claudin-1 was dramatically affected by YM201636 treatment, resulting in extensive internal accumulation of the protein (B, middle panel) compared with vehicle control (A, middle panel). Arrows highlight internal claudin-1. Arrowheads highlight claudin-1 that remained at the junctions. Scale bars 10 µm.
Figure 2.
YM201636 blocks the recycling of claudin-1.
A surface biotinylation assay was performed on control or YM201636 treated MDCK cells. (A) The ‘Surface Biotinylated’ lane represents the initial biotinylated claudin-1 at the cell surface. Labelled claudin-1 that is internal after 60 min and is resistant to surface stripping is represented for control (‘Endocytosis 60 min’) and YM201636 treated (‘Endocytosis + YM201636) cells. Claudin-1 remaining internal (and thus not recycled back to the surface) following recycling for 20 min is represented as ‘Recycling 20 min’ for control and ‘Recycling + YM201636’ for treated cells. The recycled cargo is then determined as the reduction in signal in the recycling lanes compared to the endocytosis lanes. Addition of YM201636 blocked claudin-1 recycling and caused accumulation of endocytosed protein. Our previous study shows that degradation of claudin-1 does not occur over this timeframe [10] so a block in degradation does not account for the accumulation of internal claudin-1. (B) Results represented graphically show the mean from 3 independent experiments, error bars are SEM.
Figure 3.
Claudin-2 is constantly recycled in MDCK cells and this recycling is inhibited by YM201636.
(A) MDCK cells treated with YM201636, but not DMSO controls, showed accumulation of internal claudin-2 which colocalised with the internal claudin-1 (arrows). Some claudin-2 remained at the junctions (arrowheads). Scale bars 10 µm. (B) A surface biotinylation assay was performed on control or YM201636 treated MDCK cells and a Western blot from a representative experiment is shown. (C) Results represented graphically show the mean from four independent experiments, error bars are SEM. Claudin-2 was found to be endocytosed (endocytosis 60 min) and recycled (shown by the reduction in the ‘Recycling 20 min’ lane compared to the ‘Degradation Control’ lane). Degradation would be shown by a reduction in the ‘Degradation Control’ compared to the ‘Endocytosis 60 min’ but the mean from four experiments showed there was no detectable degradation over this time scale (shown graphically in C). Addition of YM201636 inhibited claudin-2 recycling and caused accumulation of endocytosed proteins.
Figure 4.
Claudin-4 does not accumulate following YM201636 treatment and undergoes negligible endocytosis.
(A) MDCK cells treated with YM201636 showed no detectable accumulation of internal claudin-4. Scale bars 10 µm. (B) A surface biotinylation assay was performed to measure the endocytosis of claudin-4 and a Western blot from a representative experiment is shown. (C) Results represented graphically show the mean from three independent experiments, error bars are SEM. Claudin-4 was successfully biotinylated (Surface Biotinylated) but negligible claudin-4 was internalised (Endocytosis 60 min) suggesting that claudin-4 is endocytosed at a much lower rate than claudin-1 and claudin-2.
Figure 5.
YM201636 impairs tight junction formation following a calcium switch.
(A–C) Cells were stained for junctional proteins following calcium removal and showed a loss of staining at the junctions (left panels). Calcium was reintroduced in either control (middle panels) or YM201636 containing media (right panels) and cells were stained for junctional proteins. All proteins showed recovery to the junctions in control conditions. However, following YM201636 treatment claudin-1 and to a lesser extent ZO-1 failed to fully return to the junctions. (D) The transepithelial resistance (TER) of MDCK cells following calcium switch was determined. Treatment with YM201636 caused a delayed recovery of TER. Graphs represent the averages from 3 independent experiments (plus 3 replicates per experiment) and error bars show SEM.