Transport of MT1-MMP along microtubules is driven by dynein and kinesins. (A-I) Dynein siRNA leads to peripheral accumulation of MT1-MMP and to Golgi dispersal. Confocal micrographs of primary human macrophages expressing MT1-MMP–mCherry (A,E), transfected with control siRNA (A-D) or dynein siRNA (E-H), and stained for Golgi markers GM130 (B,F) and TGN46 (C,G). (D,H) Merged images. Note peripheral accumulation of MT1-MMP–mCherry (E) and dispersed appearance of Golgi (F-G) upon dynein siRNA treatment. (I) Evaluation of MT1-MMP–mCherry distribution upon transfection with control siRNA or siRNA against dynein. For each value, 3 × 30 cells were evaluated. Note enhanced percentage of cells with peripheral accumulation of MT1-MMP–mCherry in case of dynein siRNA treatment. Asterisk indicates value significantly different from controls (P < .05). For specific values, see supplemental Table 1. (J-N) Retrograde movement of MT1-MMP depends on dynein. Still images of confocal time-lapse videos of macrophages expressing MT1-MMP–mCherry and coexpressing GFP (J) or dynamitin-GFP for inhibition of dynein (K-L; supplemental Video 6), as indicated in detail images. Note pronounced peripheral accumulation of MT1-MMP–mCherry upon dynein inhibition in (K-L), mostly accompanied by reduction (K) or disappearance (L) of the central accumulation (supplemental Video 6). (M) Confocal micrograph of macrophage expressing dynamitin-GFP and stained for β-tubulin by specific primary and Alexa 488–labeled secondary antibody. Note intact microtubule system in case of dynamitin-GFP overexpression. (N) Evaluation of intracellular MT1-MMP–mCherry distribution upon overexpression of dynamitin-GFP or GFP control. For each value, 3 × 30 cells were evaluated. Note increased percentage of cells with peripheral accumulation of MT1-MMP–mCherry upon overexpression of dynamitin-GFP. **P < .001 compared with control. (O-S) Anterograde movement of MT1-MMP depends on KIF5B and KIF3B kinesins. Confocal images of macrophages expressing MT1-MMP–mCherry and coexpressing dynamitin-GFP, as indicated in detail images. Cells were treated with siRNA against control sequence (O), KIF5B (P), KIF3B (Q), or KIF1C (R). Note pronounced peripheral accumulation of MT1-MMP–mCherry upon dynein inhibition in control cells treated with control siRNA (O). This effect is inhibited by treatment with siRNA directed against KIF5B (P), KIF3B (Q), but not KIF1C (R). (S) Evaluation of cells showing peripheral accumulation of MT1-MMP–mCherry upon overexpression of dynamitin-GFP and treatment with various siRNAs. For each value, 3 × 30 cells were evaluated. **P < .001 compared with control. White scale bars indicate 10 μm.