Figure 1.
MRT cell lines are sensitive to pharmacological FGFR inhibition.
(A) Global compound selectivity analysis of MRT lines. Comparison of the three MRT lines A204, G401 and G402 versus other soft tissue cancer lines from the CCLE with regards to sensitivity to a panel of approximately 2000 compounds with defined target specificity. Shown are the top 5 enriched target in MRTs according to Activity Area and Inflection Point scores. FDR, false discovery rate. (B) Sensitivity towards the FGFR inhibitor NVP-BGJ398 among soft tissue cancer lines. A cut-off value of 500 nM was used to determine NVP-BGJ398 sensitivity based on Crossing Point values from high-throughput cell proliferation assays. (C) Immunoblot analysis of p-FRS2 and p-ERK1/2 in MRT lines treated with DMSO or NVP-BGJ398 for 40 min as indicated. Total ERK1/2 and β-Tubulin expression was used to monitor equal loading.
Figure 2.
FGFR expression levels of the MRT cell lines A204, G401 and G402.
(A) Scatter plot showing expression and copy number levels for FGFR1 (left panel) and FGFR2 (right panel) within the CCLE. MRT lines A204, G401 and G402 are indicated in red. (B) Quantitative RT-PCR (qRT-PCR) analysis of FGFR1 and FGFR2 mRNA expression in MRT cell lines and soft tissue cancer lines SKLMS1 and SKUT1. Expression values are given as average with standard errors of the mean (SEM) (n≥3) with respect to GAPDH mRNA levels (arbitrarily set as 100). (C) qRT-PCR and immunoblot analysis of SNF5-deficiency in MRT lines. SKLMS1 and SKUT1 cells were used as positive controls for SNF5 expression. SNF5 mRNA expression is given as average with SEM (n≥3) with respect to GAPDH mRNA levels (arbitrarily set as 100). β-Tubulin expression was used to monitor equal loading.
Figure 3.
Re-expression of SNF5 in MRT lines abrogates FGFR expression.
(A) Immunoblot analysis (left panel) of FGFR2 protein expression in G401 cells five days post retroviral transduction of SNF5. β-Tubulin expression was used to monitor equal loading. FGFR2 mRNA levels (right panel) were determined by qRT-PCR. Expression is shown as relative levels to control infected cells. Data are given as average with SEM (n = 3). FGFR2 mRNA expression values were normalized to GAPDH mRNA copies. Data were compared by unpaired Student’s t test; *p<0.05. (B) Analysis of FGFR1 protein and mRNA expression in G402 cells upon re-expression of SNF5 as described in (A).
Figure 4.
SNF5 loss of function induces FGFR2 expression in human fibroblasts.
(A) Effect of siRNA-mediated knockdown of SNF5 on FGFR2 expression in BJ cells. SNF5 and FGFR2 expression levels were analyzed by qRT-PCR at 72 h post siRNA transfection. Expression is shown as relative levels to cells transfected with non-targeting control siRNA and is given as average with SEM (n≥3). Expression values were normalized to GAPDH mRNA copies. (B) Immunoblot analysis of FGFR2 expression upon knockdown of SNF5 in BJ cells as described in (A). β-Tubulin expression was used to monitor equal loading. (C) Effect of siRNA-mediated knockdown of BRG1 on FGFR2 expression in BJ cells as described in (A).
Figure 5.
SNF5 is recruited to the FGFR2 promoter in BJ cells.
(A) Schematic overview of the human FGFR2 promoter. Amplicons of primer pairs used for ChIP are shown as red squares and location is indicated relative to the transcriptional start site (TSS, +1). Exons are shown as blue boxes. (B) FGFR2 promoter occupancy by SNF5 in BJ cells. Fold enrichment from chromatin immunoprecititaions (ChIP) with a SNF5-specific antibody compared to an IgG control was analyzed by qPCR using the primer pairs indicated in (A). (C) Fold enrichment of the negative control locus IGX1A and the promoter region of the known SNF5 target gene CDKN1A. Fold enrichment is given as average with SEM (n≥3). Data were compared by unpaired Student’s t test with respect to the fold enrichment of the IGX1A locus; *p<0.05.
Figure 6.
FGFR inhibition with NVP-BGJ398 impairs MRT growth in vivo.
(A) In vivo efficacy of NVP-BGJ398 in a primary mouse MRT allograft model. MRT bearing nude mice received vehicle or NVP-BGJ398 at 50 mg/kg body weight for 13 consecutive days and tumor volumes were monitored. Date are given as average with SEM (n = 4) and were compared by unpaired Student’s t test; *p<0.05. (B) Immunohistochemistry (IHC, upper panel) and immunoblot (lower panel) analysis of p-ERK1/2 levels in MRT allograft samples from mice treated with a single dose of vehicle or NVP-BGJ398 (50 mg/kg body weight) for 2 h. β-Tubulin expression was used to monitor equal loading.
Figure 7.
FGFR1 and FGFR2 expression in human primary MRTs.
Scatter plot showing expression of FGFR1 (left panel) and FGFR2 (right panel) versus SNF5 levels among human primary sarcoma samples. MRTs are indicated in red. Dashed line indicates threshold of 10th percentile of highest FGFR1 or FGFR2 expression in sample set.