ZFIN is incorporating published figure images and captions as part of an ongoing project. Figures from some publications have not yet been curated, or are not available for display because of copyright restrictions.

BMP Signaling Gradients in the Pectoral Fin

(A–C) Fin of double transgenic BRE:GFP (green) and Histone2b-mCherry (red), at 60 hpf. The BRE reporter expresses GFP under the control of Smad1/5/8 enhancers from the mouse Id1 promoter (Collery and Link, 2011, Laux et al., 2011). Dashed line, fin boundaries. Anterior, left; distal, down.

(B) Region of interest (ROI, blue; with ROI midline) where gradients are deployed: endoskeletal disc, abutting the fin fold. Cartoon indicates fin axes and endoskeletal disc (ed).

(D) BRE:GFP gradient along the straightened ROI (blue; anterior left) from (B). Orange lines indicate position s=0, corresponding to intensity maxima. Length L (black line), distance between each peak signal and the ROI midpoint (dashed line).

Scale bars: 50 μm (A–C), 20 μm (D).

(E) BRE:GFP intensity profile along the ROI midline from (B)–(D), at 60 hpf. Arrowheads, intensity maxima. Intensity corresponds to signal average orthogonal to midline.

(F and G) Anterior (F) and posterior (G) intensity profiles from (E) versus position s, with respective decay lengths (λ, slope), ± SEM. Here s=0, position of peak signal as indicated in (D)–(E). Note that the gradient profile corresponds to different levels of signaling per cell and the observed gradient is not due to different cell density or different number of signaling cells (see Figures S2K–S2M). Also, the gradient does not reflect different durations of signaling or long perdurance of GFP in the BRE reporter, since Phospho-Smad1/5/9 immunostainings show similar graded distributions (Figures S1J–S1L). Black lines, exponential fits with respective goodness of fit (R²). BRE:GFP transgene used: BRE:eGFP (Laux et al., 2011).

See also Figures S1 and S2.

Pectoral Fin Growth

(A–F) Scanning electron micrographs of larvae (A–C) and fins (D–F) at different times. Cartoon indicates fin orientation.

(G–I) DAPI (blue) and Phospho-histoneH3 immunostainings (red) at different times.

(J–L) Overlay of fins expressing zGeminin. LUT shows number of overlaid zGeminin⁺ nuclei per pixel. Dashed line, fin boundary. Distal, down; anterior, left.

(M) PD and AP fin lengths from SEM images.

(N) Number of phospho-histoneH3 cells per 10,000 μm² at different times. Line, exponential fit with goodness of fit (R²).

(O) Log-log plot of AP versus PD lengths to determine anisotropy ε (slope value) ± SEM. Line, power-law fit with goodness of fit (R²).

Black dots, average from developmental time bins. Mean ± SEM are shown in all graphs. n represents total number of fins analyzed/overlaid. Scale bars, 100 μm (A–C), 50 μm (D–L).

Scaling of BMP Signaling Gradients

(A and B) BRE:GFP gradients at different times as a function of the distance to intensity maxima (s=0, ROI midline). Bold lines, exponential fits.

(C and D) Decay length versus ROI half-length (L) of BRE:GFP gradients. Blue dots, individual data; black dots, average from length bins ± SEM. Slope (φ) values ± SEM are shown. Lines, linear fits with goodness of fit: φAR2 = 0.58; φPR2 = 0.55.

(E and F) Average dynamics of gradient amplitude (C0; SPIM imaging). Orange, growth phase (48–78 hpf).

(G) Log-log plot of normalized average amplitude C0(t)/C0(45hpf) versus normalized endoskeletal disc PD length L(t)/L(45hpf), to estimate q (slope value), with SEM. Power-law fit with goodness of fit (R²).

Shadowing corresponds to SEM per time point (E, F) or length (G). n represents number of fins analyzed. BRE:GFP transgene used: BRE:eGFP (Laux et al., 2011) (A–D); BRE-AAVmlp:eGFP (Collery and Link, 2011) (E–G). See also Figures S2 and S4.

Smoc1 Is Repressed by BMP Signaling

(A–C) BRE:GFP signal (green) and Smoc1 immunostaining (red) in the fin, at 48 hpf.

(D–G) Smoc1 expression pattern at 2 h post-heat-shock induction (hpHS) of a dominant-negative BMP receptor transgene (dnBmpr-GFP⁺) or in sibling controls (for heat-shock conditions, see STAR Methods), at 48 hpf. Inset, GFP immunostaining to monitor expression of transgene.

(H) Average intensities of normalized Smoc1 immunostainings and BRE:GFP signal versus relative position (ROI midline), at 48 hpf. Intensities were normalized to respective maxima.

(I) Average intensity of normalized Smoc1 immunostainings versus relative position (ROI midline) in dnBmpr-GFP⁺ transgenics and siblings control. Smoc1 intensity in dnBmpr-GFP⁺ was normalized to the control maxima. Shadowing corresponds to SEM per relative position (H, I).

Scale bars, 50 μm. n represents number of fins analyzed. Anterior, left; distal, down. BRE:GFP transgene used: BRE:eGFP (Laux et al., 2011) (A–C, H). See also Figure S3.

Smoc1 Mutants Scaling Phenotype

(A) smoc1 genomic structure. Exon colors match the protein domains they encode: SP, signal peptide; FS, follistatin-like domain; Thy, thyroglobulin-like domain; EC, extracellular domain. Target sequence of the Smoc1^ug104 CRISPR mutant is located in exon 4; mutant sequence of Smoc1^ug104, highlighted in magenta. Smoc1^ug104 results in a truncated protein, which lacks the first Thyroglobulin domain and other downstream features. The position of the splicing morpholino and Smoc1 antibody epitope region are indicated.

(B and C) BRE:GFP gradients in Smoc1 homozygote mutants (C) and heterozygote controls (B) displayed as straightened ROI, as in Figure 1D (anterior, left).

(D and E) Decay length versus ROI half-length (L) of BRE:GFP gradients in Smoc1^−/− mutants (black) and wild-type controls (green, from Figures 3C and 3D). Dots, average from length bins ± SEM (for individual data points, see Figures S4W–S4X). Lines, linear fits to individual data. Slope (φ) values ± SEM are shown.

(F) Comparison of φ=λ/L from BRE:GFP gradients of wild-type, Smoc1^−/− mutants and Smoc1^−/− mutants injected with smoc1 mRNA (“rescue”).

(G–I) Scanning electron micrographs of wild-type (G), Smoc1 heterozygote control (H), and Smoc1 homozygote fins (I).

(J and K) Average PD (J) and AP (K) fin lengths at 78 hpf in different mutant and morphant (MO) conditions.

(L) Comparison of average goodness of fit, R2(R‾2) obtained from collapse analysis (see Figures S4A–S4H) of BRE:GFP gradients in wild-type or Smoc1^−/− mutants. n represents number of fins analyzed.

For all statistical analyses: ^∗∗∗p < 0.0001; ^∗∗p < 0.01, ^∗p < 0.05; two-tailed, unpaired, non-parametric Mann-Whitney tests. Mean ± SEM are shown in all graphs. Anterior, left; distal down. Scale bars, 50 μm. BRE:GFP transgene used: BRE:eGFP (Laux et al., 2011) (B–F, L). See also Figures S4 and S5.