JCS02639 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure 1 -

  Fig. S1. Molecular characterization of junctional actin and thin bundles.  Keratinocytes induced to form junctions for 30 minutes were fixed and double labelled with phalloidin and different actin binding proteins (a-catenin, tropomyosin and ezrin).  Tropomyosin localized exclusively at thin bundles (arrowheads), whereas after pre-extraction, a-catenin primarily labelled junctional actin. By contrast, ezrin localized to both actin populations.

• Supplemental Figure 2 -

  Fig. S2. Sequestration of actin monomers by latrunculin affects actin recruitment and not clustering of cadherin receptors. (A) Keratinocytes grown in low calcium medium were incubated with beads coated with BSA (control) or anti-cadherin antibodies for 20 minutes in the presence or absence of latrunculin.  Unbound beads were washed away, cells were fixed and stained for phalloidin and a-catenin.  Arrows indicate recruitment of actin and a-catenin around the anti-cadherin beads.  Arrowheads indicate attached BSA-coated beads.  (B) Quantification of latex beads experiments. The ability of attached beads to recruit actin (phalloidin staining) was quantified.  Negative controls (BSA-coated beads) define the base line (25% actin recruitment).  Experiments were performed in the presence or absence of extracellular calcium ions.  Values represent the proportion of attached beads that shows actin recruitment at their periphery in four (Std Ca²⁺) or three (low Ca²⁺) independent experiments.  Error bars show standard deviation.

• Supplemental Figure 3 -

  Fig. S3. Titration of jasplakinolide dose for longer term treatment required to assess actin filament bundle formation (see Fig. 4 in the manuscript). Keratinocytes without cell-cell contacts were induced to form junctions for 60 minutes in the absence or presence of jasplakinolide at different concentrations. Cells were fixed and stained with anti-actin antibody. Higher doses (0.1 mM, 60 minutes), induced formation of actin aggregates in the cytoplasm, indicating that, under these conditions, jasplakinolide had promoted actin polymerization and general perturbation of the actin cytoskeleton as described previously (Cramer, 1999).  At lower doses (0.01 mM, 60 minutes), aggregates were not detected (compare with control; methanol). To assess whether jasplakinolide blocks actin depolymerization at 0.01 mM, we assessed lamellipodia in sub-confluent cell areas at the edge of monolayers.  In vehicle (methanol only)-treated cells, lamellipodia are highly ruffled (methanol, asterisks).  By contrast, in jasplakinolide-treated cells, we observed two effects consistent with a block in actin filament depolymerization (Cramer, 1999): lamellipodia were fragmented or completely retracted (arrowheads) and there was an accumulation of actin filament (arrows) towards the cell periphery, behind retracted lamellipodia.