Fig 2: Agent-based modeling predicts an essential role for Vangl2 in Wnt-mediated tissue patterning during zebrafish gastrulation.A, B Comparison of Wnt protein distribution in control zebrafish embryos with zebrafish embryos with Vangl2/Wnt8a overexpression. State of one realization of the model at the indicated time point (t): (Sphere stage = 4 hpf (hours post fertilization); 50% epiboly = 5.3 hpf; 80% epiboly = 8.5 hpf) for base parameter values and Wnt8a and Vangl2 over-expression parameters. Scale bar = 250 µm. For the 80% epiboly panel, the fate of the cells is also plotted in B. Source cells (S), are indicated in red. R = Receiving cells. The colors of the cells in anterior tissue correspond to the relative level of Wnt8 protein received (A). For ease of viewing, Wnt8a protein values have been normalized by the maximum value attained by any cell across the simulation and log transformed. In the cell fate diagram (B), cells acquiring a hindbrain (HB) fate are marked in dark blue, whilst those not acquiring a hindbrain fate are marked in gray. The orange line marks the estimated midbrain-hindbrain boundary (MHB). C Tissue growth properties—a single simulation. Note that the mechanics of the tissue growth are preserved across all conditions and so this graph is representative of all simulations. The red curve shows the proportion of the yolk that is covered by the tissue at the indicated times. The blue curve shows the mean tissue thickness (in terms of the number of cells). D Evolution of the cell number. E, F Histograms of cytoneme lengths at the final state of the simulation (normalized by cell diameter) for the control parameters (E) and with longer cytonemes (F). The gray curve shows a fit of the data to a log-normal distribution with means 2.0 (E) and 3.8 (F). G–K Distributions of cell fates over the angular polar coordinate at the end state of the simulation according to a hindbrain Wnt8a threshold of 100 (AU) for control parameters. The length of an arc of a circle is equal to Ø is equal to 250 µm (G), Wnt8a overexpression parameters (H), longer cytonemes (I), longer Wnt8a positive cytonemes (J), long but fewer Wnt8a positive cytonemes (K). The dark and light blue curves respectively show the mean and standard deviations of the proportion of cell fates acquiring a hindbrain fate in 100 equi-spaced bins around the yolk over 100 model simulations for each condition. The orange line shows the estimated position of the MHB. FB/MB forebrain/midbrain. The red shaded area marks the position of the margin of wnt8a producing cells. Source data are provided as a Source Data file.

Fig 3: Vangl2 regulates length and contacts of Wnt8a cytonemes in the zebrafish embryo.A, B’ Tg(vangl2:GFP-Vangl2) zebrafish embryos were injected with mem-mCherry (mCh) or Ror2-Mcherry. A’, B’ 3D renders (3D sr) of A&B at 5hpf. Yellow arrows show examples of Vangl2 positive cytonemes. A–A’ n = 6 embryos. B–B’: n = 8 embryos. C–G Wild type zebrafish embryos were injected with indicated constructs and imaged at 5hpf. Yellow arrows show examples of Wnt8a positive cytonemes. Blue arrows show examples of thick membrane protrusions. Scale bar = 10 µm. H Number of Wnt8a positive cytonemes per cell (N = number). (n per condition = 3, 6, 3, 9, 7 embryos, n = 87, 123, 44, 154, 93 cells). I Length of Wnt8a positive cytonemes (µm). (n per condition = 81, 269, 54, 82, 24 cytonemes). J Breakdown of the percentage of Wnt8a positive cytoneme lengths into 0–5, 5–10, 10–15, 20+ µm categories. K Wnt8a/Mem-mCh cytoneme with one wnt8a + ve contact point. Scale bar = 10 µm. L Vangl2/Wnt8a/mem-mCh cytoneme with multiple contact points M Quantification of number of Wnt8a-positive contact points per cytoneme. Wnt8a positive points were counted when at tips or at a fixed junction or “turning point” (n = 32, 125, 80, 81, 24 cytonemes). Graphs represent mean and standard error of the mean. Corresponding dot plots are shown in Supplementary Fig. 5. H, I, K Two-sided Kruskal–Wallis tests with Bonferroni correction for multiple tests. Statistical significance: p = 0.05. SEM = 1. Source data are provided as a Source Data file.

Fig 4: Vangl2 function is crucial for anteroposterior patterning of the zebrafish neural plate.A In situ hybridization analysis of indicated markers in 60% epiboly zebrafish embryos (6.5 hpf) injected with Wnt8a, Vangl2 and Wnt8a/Vangl2, wnt8a/vangl2/IRSp534K, wnt8a/vangl210A. Scale bar indicates 200µm. (ntl: n = 35, 16, 6, 12, 13, 19 embryos) (gbx1: n = 6, 19, 3, 11, 22, 19 embryos). a = animal pole, v = vegetal pole. Black arrowheads indicate width of gbx1 expression. Asterisks indicate ectopic expression. B Intensity of gbx1 expression from in situ hybridization (gray values in %) across the hindbrain primordium from vegetal (v) to animal (a) pole in embryos injected with indicated constructs at 60% epiboly (6.5 hpf): Control- light blue, wnt8- orange, vangl2 - gray, wnt8a/vangl2- yellow, wnt8a/vangl2/IRSp534K - dark blue. C In situ hybridization to mark pax6a expression in the primordia of the forebrain and the hindbrain in embryos injected with mRNAs for the indicated constructs at 24 hpf: control, wnt8a, vangl2, wnt8a/vangl2, and wnt8a/vangl2/IRSp534K. Scale bar indicates 100 µm. Horizontal black line indicates length of forebrain and midbrain primordium. Yellow arrowheads show the width of the hindbrain primordium, indicating the extent of convergent and extension of these cells. (n = 13, 19, 8, 10, 7 embryos). D Box and whisker plot of length of forebrain and midbrain primordia in control (blue), Wnt8a (orange), Vangl2 (gray), Wnt8a/Vangl2 (yellow) and Wnt8a/Vangl2/IRSp534K (green) 24 hpf larvae (µm). Measured from anterior forebrain pax6a expression to the position of the midbrain-hindbrain boundary (MHB) shown by horizontal black line (C). (n = 6, 12, 4, 10, 7 embryos). E Box and whisker plot of maximum width of hindbrain primordia in control (blue), Wnt8a (orange), Vangl2 (gray), Wnt8a/Vangl2 (yellow) and Wnt8a/Vangl2/IRSp534K (green) 24hpf larvae (µm). Measured from maximum width of hindbrain pax6a expression shown by yellow arrowheads (C). (n = 5, 11, 5, 8, 7 embryos). D, E Data represented as box and whisker plots. Whiskers define the minimum and maximum values. Bounds of box indicate the 25th and 75th percentile. center line indicates the median. Cross indicates mean. Outliers and inner points shown. Statistical significance: p = 0.05. D, E One-way ANOVA tests plus Tukey’s post-hoc test. SEM = 1. Source data are provided as a Source Data file.

Fig 5: Vangl2 controls the emergence of Wnt8a cytonemes in fibroblasts.A–J Wnt8a-positive PAC2 zebrafish fibroblasts transfected with indicated constructs, imaged and analysed live 24 h of post-transfection. Yellow arrows indicate examples of Wnt8a positive cytonemes. Scale bar = 10 µm. K Number of Wnt8a positive cytonemes per cell (n = number). (n per condition = 25, 9, 14, 6, 25, 14, 31, 36, 13 cells). L Length of Wnt8a positive cytonemes in PAC2 cells (µm). (n per condition = 139, 52, 32, 21, 131, 65, 47, 51, 11 cytonemes). M Breakdown of the percentage of Wnt8a positive cytoneme lengths into 0–5, 5–10, 10–15, 20+ µm categories. Graphs represent mean and standard error of the mean. K, L Two-sided Kruskal–Wallis tests with Bonferroni correction for multiple tests. Statistical significance: p = 0.05. SEM = 1. Corresponding dot plots are shown in Supplementary Fig. 2, and analysis of PAC2 filopodia is shown in Supplementary Figs. 3 and 4. Source data are provided as a Source Data file.