First published online 30 July 2003
doi: 10.1242/jcs.00693
Glycogen synthase kinase-3 regulates formation of long lamellipodia in human keratinocytes
Leeni Koivisto1,
Keyhan Alavian1,
Lari Häkkinen1,
Steven Pelech2,
Christopher A. McCulloch3 and
Hannu Larjava1,*
1 University of British Columbia, Faculty of Dentistry, Department of Oral
Biological and Medical Sciences, 2199 Wesbrook Mall, Vancouver, BC V6T 1Z3,
Canada
2 Kinexus Bioinformatics Corporation, Suite 401, 2389 Health Sciences Mall,
Vancouver, BC V6T 1Z4, Canada
3 University of Toronto, Faculty of Dentistry, Department of Biological and
Diagnostic Sciences, 150 College Street, Toronto, ON M5S 3E2, Canada
View larger version (94K):
[in a new window]
|
Fig. 1. Keratinocyte E-lam formation in mucosal wounds and in vitro. (A) Electron
micrograph of migrating keratinocytes in 3-day-old human mucosal wounds.
Arrows mark cellular extensions formed by wound keratinocytes. wb, wound bed;
kc, keratinocyte. (B) A photograph of primary mouse keratinocytes grown in
medium containing fibroblast conditioned medium and EGF (4 ng/ml). Arrows mark
E-lams. (C) HaCaT keratinocytes were treated with EGF (4 ng/ml) in DMEM
containing 1% FBS for 48 hours and then photographed. Arrows mark E-lams.
Scale bar: 10 µm.
|
|
View larger version (32K):
[in a new window]
|
Fig. 2. Staurosporine-induced E-lam formation in HaCaT keratinocytes. (A) The
morphology of HaCaT cells spreading on fibronectin in absence (a) or in
presence (b) of staurosporine as visualized by scanning electron microscopy.
Extended lamellipodia are marked with arrows and filopodia with arrowheads.
Scale bar: 10 µm. Equivalent phase contrast pictures (insets) from crystal
violet stained cells were taken using 20x objective. (B) The cells were
allowed to spread on fibronectin for 60 minutes in presence of staurosporine
(0-100 nM). The cells were fixed and four representative fields in triplicate
wells were examined by phase-contrast microscopy using a 20x objective
(n=12). The percentage of cells that were spread or forming E-lams
out of the total cell number within each field was calculated
(mean±s.d.). (C) The cells were allowed to spread on fibronectin for 2
hours and then treated with 50 nM staurosporine for 1 hour or left untreated.
For comparison, some of the cells were treated with staurosporine for 3 hours.
The cells were fixed, and the percentage of cells that were spread or forming
E-lams of the total cell number was calculated as in B (mean±s.d.,
n=12).
|
|
View larger version (123K):
[in a new window]
|
Fig. 3. The effect of tyrosine kinase and phosphatase inhibitors on E-lam
formation. The cells were allowed to spread on fibronectin for 120 minutes and
then not treated (A,B), or treated with 500 µM sodium orthovanadate (C,D),
10 µM herbimycin A (E,F) or 200 µM genistein (G,H) in the presence
(B,D,F,H) or absence (A,C,E,G) of 50 nM staurosporine for 60 minutes. The
cells were fixed, stained with crystal violet and photographed. E-lams are
marked with arrows. Scale bar: 10 µm.
|
|
View larger version (43K):
[in a new window]
|
Fig. 5. The role of GSK-3 in E-lam formation. To investigate the role of GSK-3 in
E-lam formation, the cells were treated with LiCl2 (2-40 mM; A) or
SB-415286 (1-50 µM; B) in presence or in absence of staurosporine. The
percentage of cells that were spread or representing E-lams was calculated.
Results (mean±s.d., n=8) of one representative experiment are
shown. (*P<0.05; **P<0.01;
***P<0.001.) (C) The morphology of cells treated with
50 nM staurosporine only (a) or in presence of 30 µM SB-415286 (b). E-lams
are marked with arrows. Scale bar: 10 µm.
|
|
View larger version (39K):
[in a new window]
|
Fig. 6. The localization of GSK-3 in E-lams. Immunolocalization of GSK-3 in control
(A) and staurosporine-treated (B,C) cells. GSK-3 accumulated in the extended
part of the E-lam is marked with arrows. In A and B GSK-3 was visualized by
immunofluorescence, whereas in C, GSK-3 was detected by
biotin-avidinperoxidase complex using VIP as a chromogen. To help visualize
the cell shape, partial outlines of cells are marked with dashed lines. Scale
bar: 10 µm.
|
|
View larger version (30K):
[in a new window]
|
Fig. 7. The role of intracellular free Ca2+ in E-lam formation. (A) The
spread cells were loaded with fura-2, and the ratio of 346/380 nm excitation
indicating the level of intracellular free Ca2+ was measured by
fluorimetry. Stable baseline Ca2+ was obtained for 50 seconds
before the addition of staurosporine. A typical response is shown. (B) To
study the role of intracellular free Ca2+, the spread cells were
treated with Ca2+ chelators 2 mM EGTA or BAPTA-AM (0-10 µM) in
presence or in absence of staurosporine and analyzed for cell spreading and
E-lam formation. The results show the mean±s.d. of a typical experiment
(n=8). (*P<0.05; **P<0.01;
***P<0.001.)
|
|
View larger version (29K):
[in a new window]
|
Fig. 9. The effect of Ca2+ modulators on GSK-3 phosphorylation. The
cells were allowed to spread on fibronectin for 120 minutes and then treated
with 50 nM of staurosporine alone (G,H) or together with thapsigargin (100 nM;
A,B), PLC- 1 inhibitor U-71322 (4 µM; C,D) or BAPTA-AM (10 µM;
E,F) for either 0-60 minutes or left untreated for 60 minutes (C60'). The cell
lysates were separated by SDS-PAGE and immunoblotted with antibodies
recognizing GSK-3 /ß phosphorylated on serine 21/9 (A,C,E,G) or
total GSK-3 (B,D,F,H).
|
|
View larger version (188K):
[in a new window]
|
Fig. 10. Effect of GSK-3 inhibitors on cell migration. Confluent cell layers were
scratch-wounded and then treated with staurosporine (50 nM; B,E,H), EGF (10
ng/ml; C,F,I) or left untreated (A,D,G) together with GSK-3 inhibitors
SB-415286 (30 µM; D-F) and LiCl2 (30 mM; G-I) for 24 hours. In
EGF-treated samples, the wounded cultures were preincubated for two hours
before adding EGF to the samples. The cells were then fixed, stained with
crystal violet and photographed. Scale bar: 100 µm.
|
|
© The Company of Biologists Ltd 2003
