Fig. 3
From: Rare but specific: 5-bp composite motifs define SMAD binding in BMP signaling
5-bp spaced pGC-SBE/SBE heterocomposite motifs reveal preferred MH1 binding orientation: a–c, g Dual luciferase reporter assay displaying BMP6 responsiveness towards constructs with pGC-SBE/SBE heterocomposite motifs. Whereas composite motifs with different linker sequences originating from ID1/2/3 loci compared to a control (IDL—ID-like) GC-poor linker are all BMP6-responsive (a), the relative orientation of the non-palindromic SBE motif within the composite is indicative of BMP6 responsiveness (c). BMP6 responsiveness increases with multiplication of the ID-like (IDL) composite motif to a similar level (IDL3) of the BRE2 reporter construct (b). Data are shown as mean fold induction to unstimulated cells (gray line) in relative luciferase units (RLU) ± SD (n = 3–6 independent experiments). Statistical significance was calculated in between samples using one-way ANOVA and Tukey’s post hoc test. d EMSA experiments were performed testing the binding of human SMAD1/3-MH1 domains to differently oriented pGC-SBE/SBE heterocomposite motifs. Protein concentrations (µM) are shown on top of the EMSA. Abbreviations for the DNA oligonucleotides and dsDNA sequence are shown above. Single, double, and triple SMAD-MH1 binding to dsDNA is indicated with black triangles. e–f Schemes indicating binding mode of two SMAD-MH1 domains towards composite motif. Black arrow shows SMAD-MH1 binding based on the first bound Gua (bold). e Theoretically SMAD-MH1 domains could bind in face-to-face, face-to-back, and back-to-back orientation. f Sequence comparison of BMP-responsive and unresponsive heterocomposite motifs with indicated SMAD-MH1 binding orientation. g Dual luciferase reporter assay displaying BMP6 responsiveness towards constructs containing permutated pGC-SBE/SBE heterocomposite motifs with (CTRL) GC-poor linker. While all variations lead to a decrease in responsiveness compared to WT, 6th C/1st reverse G is the most essential. Data are shown as mean fold induction to unstimulated cells in relative luciferase units (RLU) ± SD (n = 5–13 independent experiments). Statistical significance was calculated in between samples (*) or relative to WT ctrl (#) using one-way ANOVA and Å Ãdák’s (*) or Dunnett’s (#) multiple comparisons test; *P < 0.05, **P < 0.01, ***P < 0.001, and ****/####P < 0.0001