QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online December 11, 2006
doi: 10.1242/10.1242/jcs.03304

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Long, H. A. Articles by Määttä, A. Search for Related Content PubMed PubMed Citation Articles by Long, H. A. Articles by Määttä, A.

Periplakin-dependent re-organisation of keratin cytoskeleton and loss of collective migration in keratin-8-downregulated epithelial sheets

Heather A. Long^*,, Veronika Boczonadi, Lorna McInroy, Martin Goldberg and Arto Määttä

Centre for Stem Cell Research and Regenerative Medicine, School of Biological and Biomedical Sciences, University of Durham, Durham, DH1 3LE, UK

View larger version (96K): [in this window] [in a new window]   Fig. 1. Cytoskeletal organisation at MCF-7 scratch wound edges. (A) Phalloidin-stained actin cytoskeleton 30 minutes after wounding. (B) Actin cytoskeleton 2 hours after wounding. (C) SEM image of the cell surface of a wound edge cell 30 minutes after wounding displaying numerous microspikes but only a few small lamellipodia. (D-H) Immunofluorescence staining of Keratin 8 at the wound edge. (D) Low-power overview of K8 network at wounded epithelia 2 hours after wounding. (E) High-power image of K8 at the wound edge 30 minutes after wounding. (F) K8 at the wound edge 2 hours after wounding. (G) Pseudocolour gradient representation of the staining intensity form panel E. Yellow colour marks the most intense staining that is mostly localised at a cage around the nucleus. (H) Pseudocolour gradient representation of the staining intensity from panel F showing most intense keratin staining at the wound edge. (I) Percentage of cells showing the strongly stained keratin cable immediately parallel to the wound edge 30 minutes and 2 hours after the wounding. 200 cells were scored from three independent wounds. The scale bar shown in F equals to 20 µm in A, B, E-H; 50 µm in D and 200 nm in C.

View larger version (173K): [in this window] [in a new window]   Fig. 2. Field emission scanning electron microscopy of the intermediate filament network at MCF-7 scratch wound edges. Confluent monolayers grown on silicone discs were scratch wounded. After 30 minutes or 2 hours the cells were extracted and depleted from filamentous actin. (A,B) Two representative example of the wound edge IF cytoskeleton 30 minutes after wounding. (C,D) Two examples of IF organisation at the wound edge 2 hours after the wounding. (E) High magnification of the keratin cable at the wound edge. (F) Individual chromium-coated filaments at 200,000x magnification. The scale bar corresponds to 500 nm in A-D, 175 nm in E and 55 nm in F.

View larger version (74K): [in this window] [in a new window]   Fig. 3. Periplakin expression in MCF-7 monolayers and wound edges. (A-D) Intact monolayers were stained with (A) Desmoplakin and periplakin (red and green channels and merged image); XZ-projection confirms partial co-localisation the two plakins at cell borders. Note also the distinct cytoplasmic pool of periplakin, (B) Keratin 8 and Periplakin, (C) Periplakin and ß-catenin, (D) Actin and periplakin. Note that cytoplasmic periplakin is not co-localised with either keratin or actin networks in monolayer cells. (E-G) Wound edge cells stained with (E) Periplakin (left panel at both time points) and desmoplakin (right at both time points) 30 minutes and 2 hours after wounding. Desmoplakin, unlike periplakin is retained at the free wound edge at the 30 minutes time point. (F) Periplakin (green) and ß-catenin (red) merged image, 30 minutes after wounding. (G) Periplakin (red) and Actin (green), 2 hours after wounding. (H) western blot of the subcellular distribution of periplakin (Ppl), Desmoplakin (Dpl) and Keratin 18 (K18) in Saponin-soluble (S1 pool), Triton X-100 soluble (S2 pool) and insoluble pools (P3). Quantification of the relative distribution of periplakin and desmoplakin in the fractions was performed by densitometry of four independent fractionation experiments. (I) Immunofluorescence staining of cell extracted on coverslips confirming that Triton-insoluble periplakin (left panel) and desmoplakin (right panel) are located at cell borders.

View larger version (54K): [in this window] [in a new window]   Fig. 4. Co-aggregation of keratin and periplakin in Okadaic acid-treated MCF-7 cells. (A) Periplakin (Ppl) domain constructs used in transfection experiments. HA, haemagglutinin tag; Linker; the C-terminal tail of periplakin comprising the IF-binding domain (Ruhrberg et al., 1997; DiColandrea et al., 2000); Green oval, Enhanced Green Fluorescence Protein. (B) EGFP-Ppl C construct (green) and Keratin 8 (red) in MCF-7 cells. Monolayer panel: left side; EGFP fluorescence, right side; merged image. Wound edge panel: merged image of EGFP-PplC and K8 (red), note the aggregated keratin cytoskeleton in the wound edge cell. (C) Okadaic acid-treated wound edge stained for endogenous periplakin (red) and keratin 8 (green) expression. Periplakin expression is retained at cell borders, whereas the cytoplasmic pool is concentrated in small aggregates (arrowhead) or larger inclusions (arrow). (D) Subcellular fractionation of Okadaic acid and vehicle (DMSO) treated MCF-7 monolayers. S1, Saponin soluble fraction; S2, Triton X-100 soluble fraction; P, Triton insoluble pellet. (E) Transient transfection of EGFP-PplC treated with Okadaic acid. Merged image of EGFP (green) and Keratin immunofluorescence (red). Note similar aggregates (arrowhead) and large inclusion (arrow) to those seen in panel C. (F) Transient transfection of pP-C treated with Okadaic acid. Merged image of HA-tagged Ppl linker domain (green) and Keratin immunofluorescence (red). (G) Transient transfection of pP-NR treated with Okadaic acid. Merged image of HA-tagged Periplakin without linker domain (green) and Keratin immunofluorescence (red). (H) Transient transfection of pP-C treated with SB. Merged image of HA-tagged periplakin linker domain (green) and Keratin immunofluorescence (red). (I) Still images taken at indicated time points from live cell imaging of pEGFP-PplC transfection treated with Okadaic acid.

View larger version (30K): [in this window] [in a new window]   Fig. 5. Wound closure and keratin organisation in stable cell lines expressing PPL C-terminus. (A) Immunoprecipitation and western blot of Periplakin C-terminal linker domain by HA-tag antibody. MCF, untransfected parental cells; E, empty vector transfected clone, C-ter, representative clone of the linker domain transfected cells. Right panel: western blotting of endogenous periplakin and C-terminal domain with TD2 antibody. (B) Wound closure in control and Ppl-C cell lines. Wounded monolayers grown on glass coverslips were stained with FITC-phalloidin to visualise the wound edges. (C) Quantification of the wound closure in empty vector and PPL C-terminus transfected cell clones. Confluent monolayers grown in 6-well plates were wounded with pipette tip. The wound closure was followed by photographing phase contrast microscope fields after 15 minutes, 8 and 24 hours. For each wound the width of the wound after 15 minutes was designated as 100% and the subsequent time-points in the graph show the relative width of the open wound. Mean and standard deviation of ten measurements are shown. By the 24 hours time-point the control wounds were completely closed. (D). Keratin 8 immunofluorescence at wound edges. Panel (a) empty vector transfected control cell line. Panels (b) and (c) Keratin immunofluorescence at the edges of two independent wounds of a clone expressing periplakin C-terminus. (E) Ser431-phosphorylated keratin expression at wound edges. Panel (a) control; panel (b) cell line expressing Ppl C-terminus. (F) western blotting of Ser-431 phosphorylated keratin in monolayers (m) and wounded (w, ten wounds per 100 mm cell culture dish) control and Ppl C-terminal cell clones. Keratin 18 blot of the same membrane shows that there is no difference in the total keratin expression between the cell clones or treatments. Scale bar is 100 µm in B, C, E and F and 20 µm in G-K.

View larger version (60K): [in this window] [in a new window]   Fig. 6. Downregulation of periplakin and keratin 8 expression by siRNA transfections. (A) western blotting of total cell extracts from transient transfections with control or specific siRNAs. Total cell extracts were collected 72 hours after transfection and 20 µg protein aliquots were blotted with antibodies against indicated proteins. (B) Immunofluorescence staining of MCF-7 cells transfected with scrambled control siRNA (top row), Periplakin siRNA (middle row) or K8 siRNA (bottom row) showing K8 staining (red channel on the left) Periplakin staining (green channel, middle column) and merged image (on the right). (C) Control and K=8 siRNA transfected cells stained with Desmoplakin antibody (green channel, on the left) and merged with K=8 antibody staining (red channel). Scale bar is 20 µm in both B and C.

View larger version (44K): [in this window] [in a new window]   Fig. 7. Keratin organisation in periplakin siRNA transfected wound edge cells. Transiently transfected cells that had reached confluency were wounded 72 hours after transfection. (A) K8 immunofluorescence in control transfected cells. (B) K8 immunofluorescence in periplakin siRNA transfected cells. (C) Fluorescence intensity at wound edge compared with monolayer (adjusted to 100%) in control and periplakin siRNA transfected cells. Fluorescence intensity was measured in Image J programme from raw Zeiss LSM 510 images (Merged Z-stacks) from 20 cells in three independent wounds at wound edge and in epithelial monolayer 5 to 7 cell rows away from the wound edge. (D) Cell survival after siRNA transfections. Cell survival was measured using CellTiter cell proliferation kit (Promega). Mean and standard deviation (absorbance units at 490 nm) of three measurements are shown. (E) Phase contrast microscopic images of scratch wound closure of control and Periplakin siRNA transfected MCF-7 monolayers. (F) Quantification of the wound closure in MCF-7 and HeLa cells. The width of the wounds at 20 hours time point were measured from photomicrographs and displayed as percentage of remaining wound width from the start of the assay. Mean and s.e.m. from three independent transfections are shown.

View larger version (73K): [in this window] [in a new window]   Fig. 8. Wound closure in Keratin 8 siRNA downregulated MCF-7 epithelial sheets. (A) Phase contrast micrographs of the closure of control or K8 siRNA wounds at indicated time points. Cell layers were wounded 48 hours after transfection. Note the loss of epithelial phenotype in K8 siRNA transfected cells at the 20 hours time point. (B) Keratin (red) and desmoplakin (green) immunofluorescence of a control siRNA transfected wound. (C) Keratin (red) and desmoplakin (green) immunofluorescence of a K8 siRNA transfected wound. Arrowhead indicates a group of cells with no keratin immunofluorescence that have migrated away from the cell front to wound space. (D) Keratin (red) and Periplakin (green) immunofluorescence of K8 siRNA transfected wound. (E) Magnification of the area boxed in D. Note, how cells with low keratin expression have mostly cytoplasmic localisation of periplakin. Scale bar in B-D is 50 µm and in E 17 µm.

View larger version (74K): [in this window] [in a new window]   Fig. 9. K8 knockdown and wound closure in HeLa and Panc-1 cell lines. (A) Control and K8 siRNA transfected HeLa monolayers. Red channel (left), K8 immunofluorescence, green (middle) desmoplakin immunofluorescence and merged image (right are shown). (B) Western blotting of keratin and vimentin expression in the transfected cells. HeLa cells are vimentin positive but express both keratin 8 and desmoplakin. siRNA transfection effectively knocks down K8 expression in 72 hours. (C) Panc-1 monolayer transfected with K8 siRNA and stained for K8 (red) and desmoplakin. (D) Higher magnification example of desmoplakin (green) staining at Panc 1 cell borders showing punctate desmosomal staining between K8-positive (red, left panel) but not between K8 knockdown cells (right panel). (E) Phase-contrast micrographs of scratch wound closure (20 hours time-point) in Panc-1 and HeLa monolayers transfected with control or K8 siRNA oligonucleotides. Note that the wounds in Panc-1 and HeLa monolayers were wider than in experiments with MCF-7 cells. (F) Quantification of the wound closure. Open wound at 20 hours time point was measured from photomicrographs (six measurements each from two transfections for both cell lines). Scale bar: 20 µm in A and C; 5 µm in D.