Figure 1. mafba and mafbb are required for facial lymphatic development (A and B) Confocal images of (A) face and (B) trunk lymphatics labeled with Tg(fli1a:nEGFP) (endothelial nuclei, green) and Tg(-5.2lyve1b:DsRed2) (VECs and LECs, grey) in sibling and mafbb mutants at 5 dpf. LFL, lateral facial lymphatic; MFL, medial facial lymphatic; LAA, branchial lymphatic arches; OLV, otolithic lymphatic vessel; DLLV, dorsal longitudinal lymphatic vessel; ISLV, intersegmental lymphatic vessel; TD, thoracic duct. Scale bars, 100 μm. (C) Quantification of LECs in the face and trunk based on (A) in wild-type (mafbb+/+), mafbb heterozygous (mafbb+/uq47bh), and mafbb homozygous mutants (mafbbuq47bh) at 5 dpf. Embryos per genotypes; n = 10. One-way ANOVA: ∗p = 0.0174 for mafbb+/uq47bh versus mafbbuq47bh in face LECs, ns p ≥ 0.177 for other comparisons. (D and E) Confocal images of (D) facial and (E) trunk lymphatics at 5 dpf for siblings and mafba and mafba;mafbb mutants. Transgenic markers described in (A). ∗: absent lymphatics. Scale bars, 100 μm. (F) LEC quantification at 5 dpf; embryos per genotype; n = 17 (face), n = 11 (trunk), n = 10 (muLECs). One-way ANOVA: p values for facial LECs: ns p = 0.7733 for siblings versus mafbb mutants; ∗∗∗∗p < 0.0001 for siblings versus mafba or mafba;mafbb mutants and for mafbb versus mafba or mafba;mafbb mutants; ∗∗p < 0.0047 for mafba versus mafba;mafbb mutants. p values for trunk LECs: ∗∗∗p ≤ 0.0004 for siblings versus mafba or mafba;mafbb and mafba versus mafbb mutants; ∗∗p ≤ 0.0029 for mafbb versus mafba;mafbb mutants. Other comparisons: ns p ≥ 0.9999. Kruskal-Wallis test for muLECs: ns p = 0.070. (G) Gross morphology of embryos at 5 dpf. ∗: absent structures; OV, otolithic vesicle; SB, swim bladder. Scale bars, 1 mm. ns, non-significant; stars are significant; error bars are mean ±SD. See also Figure S1.

Figure 2. mafba and mafbb regulate lymphatic-vessel morphology and function but not specification (A and B) Prox1-positive LECs (magenta) (arrows) co-labeled with Tg(fli1a:nEGFP) (green) at 56 hpf in (A) face and (B) trunk in siblings and mafbb, mafba, and double mutants. Scale bars, 50 μm. (C) LEC progenitors quantification of Prox1 and Tg(Fli1a:nEGFP) double-positive cells. Embryos per genotype; n = 10; Kruskal-Wallis test: ns p ≥ 0.5331. (D) Surface rendering of Tg(-5.2lyve1b:DsRed2) for siblings and mafbb, mafba, and double mutants from confocal z stack images at 5 dpf. Scale bars, 100 μm. (E) Quantification of rendered surfaces. Embryos per genotype; n = 10. (left) Volume Kruskal-Wallis: ∗∗∗∗p < 0.0001 for siblings versus double mutants; ∗∗p = 0.0079 for mafbb versus double mutants; ∗p = 0.045 for mafba versus double mutants; ns for all other comparisons p ≥ 0.244. (Center) Area. One-way ANOVA: ∗∗∗∗p < 0.0001 for siblings or mafbb versus double mutants; ∗∗p < 0.0025 for mafba versus double mutants; ns for all other comparisons p ≥ 0.186. (Right) Sphericity. One-way ANOVA: ns p ≥ 0.234. (F) Uptake of FITC-dextran presented as a ratio of the average pixel value of z slices inside and outside the lymphatics. siblings n = 9, mafbb n = 9, mafba n = 9, and double mutants n = 8. Two-way ANOVA: from +30 min to +3 h post-FITC-dextran injection uptake is ∗∗∗∗p < 0.0001 for siblings ∗∗∗p = 0.0003 for mafbb, ∗∗p = 0.0014 for mafba, and ns p = 0.2415 for double mutants. +3 h post-FITC-dextran injection FITC-dextran uptake: ∗p = 0.0265 for siblings versus double mutants. Other interactions: ns p ≥ 0.0626. (G) Confocal projection of LFL with FITC-dextran (cyan) and Tg(-5.2lyve1b:DsRed2) (magenta) in siblings and double mutants at 5 dpf. Images of pre-injection, +30 min and +3 h post-injection. Scale bars, 100 μm. White line: trajectory for graphs in (H). (H) Fluorescence profile of Tg(-5.2lyve1b:DsRed2) (magenta) and FITC-dextran (cyan) intersecting the lymphatics from (G). ns, non-significant; stars are significant; error bars are mean ±SD. See also Figure S2 and Videos S1, S2, S3, and S4.

Figure 3. mafba and mafbb are required for valve formation and function in the facial lymphatics (A) Confocal projection of facial lymphatics and valve (arrowheads) labeled with Tg(fli1a:nEGFP) (cyan) and Tg(prox1a:RFP) (magenta) at 5 dpf, masked with surface rendering of both channels. Scale bars: 50 and 20 μm. Dashed box: magnified images. (B) Valve-length measure from (A). Embryos per genotype; n = 15. Kruskal-Wallis test: ∗∗p ≤ 0.009 for double mutants versus sibling or mafbb; ∗p = 0.044 mafba versus double mutants; other comparisons: ns p > 0.9999. (C) (Top) Projections of the average phenotype of siblings (n = 12), mafba mutants (n = 11), and double mutants (n = 12) in Tg(prox1a:RFP) embryos at 5dpf. Arrow: lymphatic valve; ∗: its absence. (Bottom) Overlay of siblings (green) with siblings and mafba or double mutants (magenta). Arrow: valve; ∗: absent valve. Scale bar, 50 μm. (D) Surface rendering of Tg(prox1a:RFP) showing valve structure in siblings, mafba mutants, and double mutants at 5 dpf. Fontal and lateral views. Scale bar, 10 μm. (E) Confocal imaging of Tg(fli1a:nEGFP) (yellow, endothelium), Tg(prox1a:RFP) (green), and Qtracker 665 (magenta) in the facial lymphatics at 5 dpf, in siblings and mafbb, mafba, and double mutants, 5 min post-injection. FCLV, facial collecting lymphatic valve; FLV, facial lymphatic vessel. ∗: absence of dye in FCLV. Scale bar, 50 μm. (F) Percentage of embryos with dye leakage into FCLV at 5 min post-injection in siblings (n = 17), mafbb mutants (n = 8), mafba mutants (n = 15), and double mutants (n = 6). Chi-squared; p = 0.0159. Single comparisons with Fisher’s exact test: ∗∗ p ≤ p = 0.009 for double mutants versus siblings or mafbb; ∗p = 0.0456 for mafba versus double mutants; other comparisons: ns with p ≥ 0.399. ns, non-significant; stars are significant; error bars are mean ±SD. See also Figure S2 and Video S2.

Figure 4. mafba and mafbb regulate facial LEC migration downstream of Vegfc-Vegfd-SoxF

(A) (Left) Confocal images of Tg(flt4BAC:mCitrine) expression in facial lymphatics in sibling, sox7^+/−;sox18^−/−, sox7^−/−;sox18^+/−, and sox7^−/−;sox18^−/− embryos at 3 dpf. ^∗: reduced sprout length. Scale bars, 100 μm. (Right) Quantification of facial lymphatic sprout length at 3 dpf in sox7 and sox18 mutant backgrounds. sox7^+/+;sox18^+/+ n = 12; sox7^+/−;sox18^+/+ and sox7^−/−;sox18^+/+ n = 13; sox7^+/+;sox18^+/− n = 18; sox7^+/−;sox18^+/− n = 30; sox7^+/+;sox18^−/− n = 7; sox7^+/−;sox18^−/− n = 17; sox7^−/−;sox18^+/− n = 32; sox7^−/−;sox18^−/− n = 11. One-way ANOVA: ^∗∗∗∗p < 0.0001 for sox7^+/+;sox18^+/+ versus sox7^−/−;sox18^+/− or sox7^−/−;sox18^−/−; other comparisons: p > 0.9782.

(B) (Top) Heatmap of Tg(mafbbE1bas:EGFP) signal intensity from confocal projection at 2.5 dpf in face LECs. Gray value intensity scale: 0–113 for control and MO-vegfc + MO-vegfd, scale: 0–147 for control and MO-sox7+MO-sox18. White bracket: quantification area. Scale bar, 50 μm. (Bottom) Quantification of Tg(mafbbE1bas:EGFP) intensity. (Left) Control n = 7; MO-vegfd and MO-vegfc n = 9; MO-vegfc + vegfd n = 8; (right) control n = 17; MO-sox7 n = 9; MO-sox18 n = 14; MO-sox7 + sox18 n = 10. Kruskal-Wallis test: ns p > 0.2372.

(C) Quantitative real-time PCR of mafba and mafbb expression relative to cdh5 in facial LECs at 48 hpf. 12 replicates. Wilcoxon test: ns p = 0.3013.

(D and E) Quantitative real-time PCR of mafba and mafbb expression in face LECs at 48 hpf. (D, left) 6 replicates for control and MO-vegfc and 5 for MO-vegfd and MO-vegfc + vegfd; expression relative to cdh5. (D, right) 6 replicates for control, MO-sox7, 5 for MO-sox18, and 4 for MO-sox7+MO-sox18; expression relative to B actin. (E, left) 6 replicates for control, MO-vegfd, MO-vegfc, and MO-vegfc + vegfd; expression relative to cdh5. (E, right) 6 replicates for control, 5 for MO-sox7 and MO-sox18, and 4 for MO-sox7+MO-sox18; expression relative to B actin. Kruskal-Wallis: p values reported on the graph.

(F) Confocal projections from time-lapse imaging of LEC migration in siblings and double mutants labeled by Tg(-5.2lyve1b:Venus) from 40–48 hpf. Arrows: leading cells in FLS. Scale bars, 50 μm.

(G) (Left) FLS migratory distance (from F). Embryos: siblings n = 15, mafbb n = 11, mafba n = 12; double mutants n = 9. One-way ANOVA: ^∗p = 0.0279 for siblings versus double mutants; ns for all other comparisons. (Right) FLS migratory velocity (from F). Cells: siblings n = 24, mafbb n = 16, mafba n = 14; double mutants n = 29. One-way ANOVA: ^∗∗∗∗p < 0.0001 for siblings versus mafbb, and mafbb or mafba versus double mutants; ^∗∗∗p = 0.0001 for siblings versus mafba; ns for all other comparisons.

(H) FLS tip mean-squared displacement (MSD) (from F) in siblings (n = 4) and double mutants (n = 4).

(I) (Top) Schematic representation of directionality in the polar histograms. (Bottom) Polar histograms of migration tracked (from F) in siblings (n = 4) and double mutants (n = 4). Collective cells: siblings n = 24 and double mutants n = 29; leading cells: siblings n = 4 and double mutants n = 4; following cells: siblings n = 20 and double mutants n = 25. ns, non-significant; stars are significant; mean with SD.