Fig. 5

T-RECS regulates hnRNPA2/B1 protein stability. a) Immunoblotting showing a strong decrease in hnRNPA2/B1 protein levels 1-day after T-RECS ASO treatment compared to Control ASO treatment in D04 cell lysate. β-actin served as a loading control. b) Representative images of fluorescent signals received from D04 cell pellets treated with either T-RECS ASO, or Control ASO shows a strong reduction of fluorescent signal of hnRNPA2/B1-binding antibody. c) Immunoblot analysis of D04 cell lysate derived from hnRNPA2/B1 and Rabbit IgG pulldown samples confirms successful hnRNPA2/B1 pulldown. d) To account for potential unspecific binding to the RNA-binding protein hnRNPA2/B1, enrichment analysis (calculated at 10%-input) was executed by comparing T-RECS enrichment in the hnRNPA2/B1 pulldown samples to other lncRNAs. T-RECS was > 65-fold enriched, while HOTAIR (5.4-fold, p = 0.03) and MALAT1 (18-fold, p = 0.049) were significantly less enriched. e) Differential expression analysis of RNASeq data unveils that compared to T-RECS expression, the lncRNA HOTAIR is not significantly higher expressed (1.47-fold increase, p = 0.1) and lncRNA MALAT1 is significantly higher expressed (530-fold increase, p = 0.003) in the D04 cell line. f) Representative ISH-derived (RNAscope) fluorescent signals from D04 and g) MM415 cell pellets are in green (T-RECS), orange (hnRNPA2/B1 protein), or blue (DAPI-nuclear control). Nuclear compartments with high hnRNPA2/B1 expression levels tend to overlap with areas of higher T-RECS expression. h) hnRNPA2/B1 expression analysis of non-malignant skin biopsies from the GTEx dataset (n = 1305, mean TPM: 317), and BRAF- and NRAS-mutated melanoma tissue from the TCGA-SKCM dataset (n = 366, mean TPM: 430) shows that hnRNPA2/B1 is significantly upregulated (p < 0.001) in MAPK-pathway hyperactivated melanoma. The center line represents median expression, the box represents the lower and upper quartiles, the whiskers extend to the furthest value that is less than 1.5 times the interquartile range from the lower and upper quartiles and the mean expression is marked by an ‘X’. i) In healthy-skin patient samples (GTEx, n = 1305), the expression correlation coefficient of T-RECS and hnRNPA2/B1 (ρ = 0.22, red dot) is not statistically different from the coefficient of T-RECS (mean ρ = 0.08, p > 0.09, blue dot in left panel) or hnRNPA2/B1 (mean ρ = 0.12, blue dot in right panel, p > 0.26) compared to a set of 200 randomly chosen genes (gene set 1). j) In MAPK-pathway driven melanomas (TCGA, n = 366), the expression correlation coefficient of T-RECS and hnRNPA2/B1 (ρ = 0.43, red dot) is significantly higher than the coefficient of T-RECS (mean ρ = 0.1, blue dot in left panel, p < 0.003) or hnRNPA2/B1 (mean ρ = 0.12, blue dot in right panel p < 0.035) compared to a set of 200 randomly chosen genes (gene set 1). These findings are consistent with those obtained from nine additional gene sets as shown in Suppl. Figure 3c-d. Significance for panel d-e) and h) is shown as p-values calculated by Student’s t-test. * = p < 0.05, ** = p < 0.01, *** = p < 0.001. Significance for panel i-j) was evaluated by comparing the correlation between T-RECS and hnRNPA2/B1 with correlations between either T-RECS or hnRNPA2/B1 with 200 randomly selected genes and calculating a Z-score. Error bars represent standard deviation