Figure 1.
Basal FHL2 expression in human osteosarcoma cells and in tissue microarrays (TMA) of human osteosarcomas.
Whole cell lysates were probed with the indicated antibody and revealed by Western blot analysis (A). FHL2 expression was determined by immunohistochemistry in tissue sections of normal bone, primary tumors, metastatic or recurrent osteosarcoma (Mag.×125) (B). Semi-quantitative scoring of immunohistochemical staining with anti-FHL2 antibody in normal bone and osteosarcoma samples according to patient outcome (primary tumor, metastatic or recurrent osteosarcoma) (C). *P<0.05.
Figure 2.
FHL2 silencing decreases Wnt/β-catenin signaling in osteosarcoma cells.
Cell lysates of osteoblast precursor cell (C3H10T1/2), calvaria-derived osteoblastic cells (MC3T3E1) and osteosarcoma cell lines (K7M2) were analysed by western blot and FHL2 level was corrected for β-actin (A). After transduction with shControl or shFHL2, FHL2 levels in K7M2 cells were evaluated by q-PCR (B) and Western blot analysis (C). shControl and shFHL2 transduced K7M2 cells were treated for 24 h with Wnt3a CM and β-catenin nuclear translocation in K7M2 cells was evaluated by Western blot analysis of nuclear fraction (D), immunocytochemistry (red, arrows: β-catenin, blue: DAPI) (E), and β-catenin transcriptional activity was determined by a reporter assay (F). The mRNA levels in the shControl and shFHL2 cells were evaluated by q-PCR analysis (G, H). *: P<0.05 vs the indicated group or shControl cells.
Figure 3.
FHL2 silencing decreases osteosarcoma cell growth.
After treatment with Wnt3a CM (A) or FGF-2 (0.50 ng/ml) (B) for 3 days, DNA replication was evaluated by BrdU incorporation in shControl and shFHL2-transduced K7M2 cells. Apoptosis was induced by serum deprivation and effector caspases activity was evaluated at 48 h in cells treated with or without Wnt3a CM (C). a: p<0.05 vs untreated, b:p<0.05 vs shControl. TUNEL analysis was performed in basal and serum deprivation conditions at 72 h (D). *: P<0.05 vs the indicated group or shControl cells.
Figure 4.
FHL2 silencing decreases bone tumor cell migration and invasion.
Migration of shControl and shFHL2-transduced K7M2 cells was evaluated by Boyden’s chamber (A) and wounding assays (C) and migrating cell number was evaluated (B, D). K7M2 cell invasion was evaluated by the Matrigel invasion assay (E, F). *: P<0.05 vs shControl-transduced cells.
Figure 5.
FHL2 silencing decreases bone tumor growth in vivo.
shRNA control and shFHL2-transduced murine K7M2 cells were injected in BALB/c mice and tumor size (A) and volume (B) were determined at 6 weeks (n = 9 per group). Cell proliferation and apoptosis in tumors was determined by histological analysis using Ki67 (C, D) and TUNEL staining (arrows), respectively (E, F). Wnt5a and Wnt10b mRNA expression was evaluated in the tumors by q-PCR analysis (G). *: P<0.05 vs shControl cells.
Figure 6.
FHL2 silencing reduces lung metastasis in mice.
Histological hematoxylin/eosin (H&E) staining of lung tissue sections showing metastasis (stars) developed in mice injected IM with shControl or shFHL2-transduced K7M2 cells (A). Metastasis area (B) and number (C) in the lung tissue were evaluated. Results are expressed as mean ± s.d. (n = 9 animals per group). *P<0.05 vs shControl. Proposed model in which FHL2 silencing using shFHL2 in murine osteosarcoma cells attenuates Wnt/β-catenin signaling and reduces the expression of Wnt5a and Wnt10b and possibly other FHL2 target genes in the tumors, resulting in decreased osteosarcoma cell growth, invasiveness and tumorigenesis in vivo (D).