Figure 4
CD63+ vesicles typically localize near invadopodia in MDA-MB-231. (a) Representative epifluorescence images of F-actin (Alexa fluor 594 phalloidin staining) and CD63 in parental MCF-7 (top panels) and MDA-MB-231 (bottom panels) breast cancer cells seeded onto a gelatin matrix. MCF-7 cells do not form invadopodia (phalloidin, left top panel) and have a diffuse CD63-pattern (middle top panel). Conversely, MDA-MB-231 do form invadopodia which can be observed as perinuclear phalloidin dots (left bottom panel). The CD63 signal organizes in a very discrete region (middle bottom panel) and this region colocalizes with the perinuclear invadopodia region (right bottom panel). (b) Quantification of the CD63 dot density at the total cell area (left), invadopodia region (middle) and total cell area without the invadopodia region (right) in parental MDA-MB-231. Tukey boxplots represent quantification of the amount of CD63 dots per pixel. P0.05-value was determined by a Mann-Whitney rank sum test and n ≥100 cells were quantified for each boxplot. CD63 dot density is significantly higher at the invadopodia region compared to the total cell area (p < 0.0001), or compared to the cell area minus the invadopodia region (p < 0.0001), as illustrated in (a) (bottom panels). ***P ≤ 0.001. (c) Representative confocal images of parental MDA-MB-231 cells seeded onto a fluorescent gelatin matrix. Invadopodia are visualized by immunofluorescence of TKS5, CTTN and F-actin (Acti-Stain 670 phalloidin staining). Perinuclear F-actin dots colocalize with TKS5 (top panels) and CTTN (middle panels) respectively, and with areas of matrix degradation, confirming the presence of invasive invadopodia. CD63 on the other hand, occasionally colocalizes with F-actin and with areas of matrix degradation (invadopodia), but both signals are always near each other (bottom panels). Boxed areas are enlarged in the bottom right insets and arrowheads Δ indicate areas of colocalization.
