Caveolin-2 interacts with syndecan-2 and regulates cell adhesion in MDA-MB231 cells. (A) Syndecan-2, but not syndecan-4, was co-immunoprecipitated with caveolin-2 from cell lysates. Co-immunoprecipitation was not affected by ectopic glycosaminoglycan treatment of the cells (UT; untreated, CS; chondroitin sulfate, HS; heparan sulfate). (B) Confocal laser scanning microscopy and profile of the line scanning (white arrow on image) confirmed partial co-localization of syndecan-2 (green) and caveolin-2 (red). (C) Caveolin-2 levels were reduced where syndecan-2 was depleted by siRNA, compared with control siRNA. (D) Western blotting verified the knockdown efficiency of siRNA targeting caveolin-1 and −2 compared to control siRNA. Downregulation of caveolin-1 reduced the expression of caveolin-2 by around 30%, but knockdown of caveolin-2 had no impact on caveolin-1 levels. (E) F-actin containing microfilament bundles were abundant after caveolin-2, but not caveolin-1 depletion. Bar = 25 μm. (F) Spread cell areas were measured in caveolin-2 depleted cells and control cells, n ≥ 50 per condition. (G) Adherens junctions and focal adhesions (arrows) were characteristic of caveolin-2 depleted cells, shown by cadherin-11 and p120-catenin, or paxillin distributions. Bar = 25 μm. (H) Microfilament bundles formed in response to either syndecan-2 or caveolin-2 knockdown were sensitive to 30 μM Rho kinase inhibitor, Y-27632. Bar = 25 μm. (I) Diphosphorylated myosin light chain (ppMLC; Thr18/Ser19) was enhanced in both syndecan-2 and caveolin-2 depleted cells. Densitometry analysis of western blots for ppMLC and total MLC were normalised to β-tubulin. Similar results were obtained in three independent experiments. (J, K) Caveolin-2 depleted cells had reduced ability to invade or degrade type I collagen gels. Higher magnification images are shown in the lower panels. Caveolin-1 depletion also reduced collagen gel invasion. Bar = 100 μm. Error bars = s.e.m. **p < 0.01, ***p < 0.001, n.s.; not significant by two-tailed paired t-test.