Figure 4

CAPNb2 restrains lung metastasis of breast cancer cells. (A) In vitro cell migration. CAPNb2 significantly reduces migration of MDA-MB 231 cells; n = 6 (CAPNb2), 7 (CAPNb4) and 7 (GFP), P = 0.017 CAPNb2 versus GFP. Ns, not significant. (B) Matrigel invasion. MDA-MB-231 cells stably expressing CAPNb2 are not invasive compared with GFP-transduced cells (t test, P = 0.046; n = 3). RFU, relative fluorescence unit. (C, D) MDA-MB-231-CAPNb2-EGFP cells display fewer broad lamellipodia (t test, P = 0.00013; n = 3). (E) Orthotopic metastasis model. Noninvasive bioluminescence images of mice injected with GFP control cells or CAPNb2 cells. A red line (55d) marks the region of interest. H/E stainings of lung sections are at right. Arrows, macrometastases. (F) Normalized photon-flux profile from lungs at the indicated times in the orthotopic metastasis model. t₃₈P = 0.0240; t₅₅P = 0.0339; n = 6 to 7. (G) Tail-vein metastasis model. Bioluminescence images of mice injected with GFP control cells or CAPNb2 cells. A red line (55d) marks the region of interest. H/E stainings of lung sections are at right. Arrows, macrometastases. (H) Normalized photon-flux profile from lungs at the indicated times in the tail-vein metastasis model. t₁₈P = 0.0204; t₃₈P = 0.0092; t₅₅P = 0.0104; n = 7. A mouse in the GFP cohort (n = 4) with an unusually high normalized photon flux of 2.93 × 10⁷ photons/s was excluded. The total photons/s was measured in a standard-sized circular region of interest encompassing the murine chest. For normalization, we set to 1 the average photon flux at day 0 of the lungs of mice injected with CAPNb2-GFP cells. Error bars represent mean ± SEM except for D, in which mean ± SD are shown. *P < 0.05; ***P < 0.001.