Fig. 4.

Integrin adhesion is required for stress-fibre organization and the control of F-actin content at the basal surface. (A-F) Micrographs of clones lacking βPS, with clones marked either by the absence of GFP (A,B; no green) or the presence of GFP (C-F; green). (A′,B′) Show actin distribution; (A,B) show the merge of actin (red) and the position of the clone. (A,A′) Stage 6 egg chamber showing mutant cells forming thicker filaments than wild-type neighbours. (B,B′) Basal F-actin is stronger in cells lacking βPS at stage 12. (C,C′) In the absence of integrin, basal F-actin fails to assemble into stress fibres (C) and cells fail to flatten as indicated by nuclei at a lower focal plane (C′). (D) A mid-cell section shows the rounded shape of the cells lacking integrin (actin, red; nuclei, blue); the yellow lines show the shape of wild-type cells. (E,F) Transverse sections of follicular epithelia [actin, red; nuclei, blue; CD8-GFP, green (mutant cells); actin channel, grey). Red bars on the right indicate the thickness of the epithelia and apical (`A') and basal (`B') surfaces. The oocyte membrane is just above `egg'. Green asterisks on the actin channel indicate the position of mutant cell(s). Cells lacking integrin were rounded and failed to flatten at stage 12 (E), whereas stage 10A cells seemed normal (F). (G) Quantification of basal F-actin in cells lacking integrins (βPS) compared with wild type (WT) measured as integrated area (grey mean value multiplied by cell area). Wild-type value (92) divided by mutant value (122) is 0.75, which indicates a 33% increase in basal actin in the mutant cells. P=2.8×10^–5 (Student's t-test two samples assuming equal variance). Scale bars: 20 μm. Scale bar shown in A is for A,A′; scale bar shown in B is for all other micrographs except D.