Supplementary Figure 3: ERK1 and ERK2 influence on primary tumours and metastasis.

(a) ERK1 and ERK2 downregulation was confirmed by Western blotting. (b) Percentage of mice intrasplenically injected with LiM2 cells expressing the indicated shRNAs that developed liver metastases (up) or lung metastasis (bottom). n = 17 (shControl), n = 10 (sh1 ERK1) and n = 18 (sh2 ERK1) number of mice used per group pooled from two independent experiments. Statistical analysis was done using two-sided Fisher’s exact test. n.s.-not significant. (c) LiM2 cells expressing the indicated shRNAs were injected orthotopically into mice cecum and growth rates by photon flux quantification were determined. n = 9 (shControl), n = 9 (sh1 ERK1), n = 8 (sh2 ERK1), n = 10 (sh1 ERK2), n = 8 (sh2 ERK2) number of mice used per group. Data is presented as average plus s.d. Statistical analysis was done using one-way ANOVA and no statistical differences were found. (d, e) Size matched lesions from (c) were selected for immunohistochemistry Ki-67 (d) and cleaved caspase-3 staining (e). n = 20 total fields per group analysed where five different fields from size matched lesions per mouse were quantified. Number of mice used in each group was 4. Statistical analysis was done using two-tailed Mann Whitney test. All values are presented in the graphs together with average ±s.d. (f) Kaplan–Meier curves representing association of proportion of recurrence-free patients with relative FRA1 gene expression levels in human primary colon cancer dataset (pooled GSE17537 and GSE14333; n = 267). HR-hazard ratio. Statistical analysis was done using Cox proportional hazards model.