Figure 6: Functional importance of Roquin-1 target motifs in cells.

(a) Overview of the Ox40 3′-UTR and truncated/mutated versions thereof as used for EMSA assays in b and the expression experiments of Ox40 in c and d. (b) EMSA experiments probing the interaction between the Roquin-1 N-terminal region (residues 2–440) and either the complete wild-type Ox40 3′-UTR or versions with mutations of the CDE-like SL, the ADE-like SL or both SLs (see a). Arrows indicate the individual binding events to either motif. It is noteworthy that the higher bands observed at large protein concentrations are probably additional nonspecific, lower-affinity interactions of Roquin-1 with the 3′-UTR or protein aggregates. (c) Relative Ox40 MFI normalized to expression levels from the Ox40 CDS construct. Error bars show s.d. of seven (CDS, 1–40, 1–80, 1–120 and full-length), six (ADE-like mut and CDE mut) or three (double mut) independent experiments. Statistical significance was calculated by one-way analysis of variance (ANOVA) Kruskal–Wallis test followed by Dunn’s multiple comparison test (**P<0.01). (d) mRNA decay curves of Hela Tet-Off cells stably transduced with retroviruses expressing Ox40 CDS without 3′-UTR (CDS, red line), Ox40 CDS with its wild-type 3′-UTR (full length, black line), Ox40 full length with mutated ADE-like motif (ADE-like mut, grey line), Ox40 full length with mutated CDE-like motif (CDE-like mut, green line) or Ox40 full length with mutated ADE and CDE motifs (Double mut, blue line). Error bars represent the mean of technical duplicates in one experiment. mRNA half-life times were calculated with Graph Pad Prism. Data are representative of two experiments with similar results.