Fig 2: Identification and verification of E2F recognition elements in the chick Atoh1 conserved elements. (A) Schematic diagram showing the location of nine putative E2F binding sites predicted in the chick Atoh1 conserved regions by MatInspector software (labelled as S1–S9) Numbers indicate the position of the enhancers in base pairs. (B) Immunofluoresence detection of endogenous expression of ATOH1 (green) and E2F1 (magenta) proteins in UB/OC-2 cells co-stained with DAPI (blue). (C) and (D): EMSA analysis performed using a radiolabelled E2F1 consensus sequence, containing a functional E2F1 binding site incubated alone (lane 1), with untransfected nuclear extracts or with nuclear extracts transfected with E2F1 and DP1 expression vectors as indicated. Competition assays were performed with 500 ng of non-radiolabelled competitors (consensus E2F1, lane 3 in both (C) and (D), and lane 9 in (C); mutant E2F1, lane 4 in (C) and (D), and lane 10 in (C); non-specific (N/S) probe, lane 5 and 11 in (C); sites S1 to S9 as described in (A), lanes 7–15 in (D). Lane 6 and 12 show a supershift assay with rabbit polyclonal anti-E2F1, pre-incubated with extracts before the binding assay.

Fig 3: Effect of the E2F transcription factors on the mouse and chick Atoh1 enhancers and putative enhancer C. (A) Schematic representation of the luciferase constructs containing the mouse or chick Atoh1 conserved regions cloned into the luciferase vector pGL4.23. (B) Dose–response effects on the Atoh1 luciferase constructs of transfecting increasing amounts of an E2F1 expression construct or mutant E2F1 in the UB/OC-2 cell line. (C) Effect of E2F2, E2F3 and E2F4 on the activity of the Atoh1 luciferase constructs. Experiments were conducted in triplicate in two separate assays with different DNA preparations for each plasmid in each assay (n = 6). Relative luciferase values are expressed relative to activity of the reporters in the absence of E2F. Student t-test was conducted comparing each condition with 0 ng of the corresponding E2F expression construct. (*p < 0.05; **p < 0.01; #p < 10–5; ##p < 10–10). Error bars represent the s.e.m (n = 6).

Fig 4: E2F1 expression during HC regeneration in the chick BP. (A) Schematic representation of the experimental timeline of organotypic cultures of E18 chick BPs cultured in DMEM and 1% FBS for 2 days with 78 µM streptomycin (Strep) followed an additional 3 or 6 days in vitro (DIV) in the same media without streptomycin. The arrows indicate when the tissue was harvested. (B) Immunofluorescence of BPs showing the expression of E2F1 and ATOH1 in HCs (labelled with Phalloidin) and SCs (labelled with SOX2). Cultures maintained for 2 days in media with Strep show ATOH1 re-activation in some SCs whereas a downregulation of nuclear E2F1 expression is observed in comparison to control cultures maintained in DMEM (hollow arrowheads in zoom in areas in row (I) and (IV); scale bar: 10 µm. In BPs cultured for 3 additional days in streptomycin-free media (2d Strep + 3 DIV), nuclear E2F1 expression is observed in SCs (hollow arrowheads in zoom in areas in row II); scale bar: 10 µm). New HCs labelled with Phalloidin are formed in cultures maintained for 6 days in streptomycin-free media after drug treatment (2d Strep + 6 days DIV). In these cultures, ATOH1 is downregulated in SCs labelled with SOX2 whereas E2F1 is expressed in SCs and HCs (arrowheads in row III in E2F1 panel). Control experiments were conducted in parallel in DMEM and at the same timings plus additional days of in vitro (DIV). Technical replicates were three for each experiment with at least three biological samples.

Fig 5: The effect of site directed mutagenesis of the E2F1 putative sites 2 and 6 on E2F1 mediated activation of the Atoh1 regulatory region. (A) Consensus seuqeunce of the E2F1 binding site described by MatInspector, Genomatix. Cross species alignment of sites 2 and site 6 in the Atoh1 regulatory regions. Conserved nucleotides within the core sequence of site 2 and site 6 are shown in red. (B) Diagram showing the sequence and position of putative E2F sites 2 and 6 which both contain overlapping predicted binding sites for E2F1 and E2F4. Point mutations were introduced in the core GC nucleotides, replaced with TT (marked in bold and with asterisks). (C) Luciferase assays show a reduced response of the chABC-s2mut (*p < 0.05) and the chABC-s6mut (***p < 0.001) to E2F1 transfection in comparison with the wild type luciferase construct (chABC-wt). Each experiment was conducted in triplicate in two separate assays with different DNA preparations. Student t-test was conducted.