Fig 2: Overexpression of the 388Arg variant of FGFR4 induces the expression of EMT markers in lung cell lines. The mRNA (A) and protein (B) expression of four EMT markers (N-cadherin, vimentin, Twist1 and Snail1) was performed in the control empty vector (EV), FGFR4-388Gly-overexpressing (FGFR4-Gly) and FGFR4-388Arg-overexpressing (FGFR4-Arg) lung immortalized NL20 cell line and lung SCC H226 and Calu1, and ADC H2009 and HCC827 cell lines. (C) Relative number of migrated cells of the previously mentioned cell lines. The mRNA measurements were performed in three independent experiments and mean expression values, represented as 2−ΔCt with their respective standard deviation, are represented. For western blots, a representative blot is shown. Migration assays were performed in three independent experiments. All values were normalized to empty vector control for each replicate of the experiment and the mean and standard deviation of every normalized replicate are represented. p-values are represented as asterisks (*p < 0,05; **p < 0,01; ***p < 0,001, ns = non significant). EV = Empty Vector control, FGFR4-Gly = FGFR4-388Gly overexpressing, FGFR4-Arg = FGFR4-388Arg overexpressing.

Fig 3: Correlation of FGFR4 variant and prognosis in high FGFR4 mRNA expressing NSCLC patients. (A) Overall and progression-free survival curves for high FGFR4 mRNA expressing patients, according to the FGFR4 variant in the whole NSCLC cohort. (B) Overall and progression-free survival analysis of SCC and ADC patient subsets depending on the FGFR4 variant, taking into account exclusively the groups with high FGFR4 mRNA expression. Gly = FGFR4-388Gly, Arg = FGFR4-388Arg.

Fig 4: Effect of the overexpression of the 388Gly and 388Arg variants of FGFR4 in the tumorigenic abilities of the inmortalized NL20, the lung SCC H226 and Calu-1, and the H2009 and HCC827 lung ADC cell lines. (A) 10% FBS growth curves. (B) soft agar assays showing the relative colony number quantification (left) and representative images (right). (C) Western blot determination of activation of cancer-related signalling pathways of 388Gly and 388Arg FGFR4-overexpressing cell lines. In soft agar assay, colony number representation is shown. All values were normalized to empty vector control for each replicate of the experiment, and the mean and standard deviation of every normalized replicate are represented. For western blotting, cells were serum starved for 5 hours and then the protein extraction was made. For the serum stimulated conditions, after serum starvation cells were stimulated with serum-containing complete medium for fifteen minutes before protein extraction. p-Values are represented as asterisks (*p < 0,05; **p < 0,01; ***p < 0,001). EV = Empty Vector control, FGFR4-Gly = FGFR4-388Gly overexpressing, FGFR4-Arg = FGFR4-388Arg overexpressing, FBS = Fetal Bovine Serum.

Fig 5: In vivo effects of the overexpression of the 388Gly and 388Arg variants in the H2009 ADC and H226 SCC cell lines. Relative tumor growth representation of the xenografts generated in immunodeprived nude mice of the reported cell lines. p-values are represented as asterisks (*p < 0,05; **p < 0,01; ***p < 0,001). EV = Empty Vector control, FGFR4-Gly = FGFR4-388Gly overexpressing, FGFR4-Arg = FGFR4-388Arg overexpressing.