Two lmx1b mutant fish lines were generated using CRISPR-Cas9. (A) Schematic showing exons targeted by gRNAs (black arrowheads) in gene and protein sequences for lmx1ba and lmx1bb. HD, homeobox domain. (B) Stereomicroscope images of wild type and lmx1b mutant lines at 5 dpf. Black arrowheads indicate oedema; white asterisks show loss of swim bladder inflation in lmx1bb^−/− and dKO; white arrowhead shows body truncation in dKO. Scale bars: 250 μm. Quantification of number of larvae at 5 dpf showing (C), oedema at varying severities and (D) swim bladder inflation or non-inflation. N=19 for wild type; 38 for lmx1ba^−/−; 19 for lmx1bb^−/− and 25 for dKO. (E) Graph showing body length measurements from 1–7 dpf in wild-type, lmx1ba^−/−, lmx1bb^−/− and dKO larvae. Two-way ANOVA performed comparing to wild type, where *** P=0.009 and ****P<0.0001 for dKO; † P=0.0458 and †† P=0.0064 for lmx1bb^−/−. (F) Kaplan–Meier survival curve showing survival rate of lmx1b mutant lines compared to wild type N=14 for wt and lmx1ba mutants; 20 for lmx1bb mutants; 36 for dKO mutants.

Loss of lmx1bb disrupts kidney formation and function. (A) Graphs showing percentage of lmx1bb larvae within each oedema classification at 5 dpf (left) and 7 dpf (right). N=9 for wild type; 16 for lmx1bb^−/+; 19 for lmx1bb^−/− mutants. (B) Schematic showing normal layout of kidney glomerulus. Electron micrographs images of the glomeruli of (C) wild type, and (D,E) lmx1bb mutants at 6 dpf. For C, pink and white arrowheads indicate interdigitating podocyte foot processes, white dashed lines show GBM, and pink asterisks indicate fenestrations. For D and E, white and purple arrowheads indicate possible interdigitating foot processes, white bracket shows region of foot processes. BV, blood vessel; EC, endothelial cell; F, fenestrations; FP, foot processes; G, glomerulus; GBM, glomerular basement membrane; gbm, partial glomerular basement membrane; IDP, interdigitating podocyte foot processes (indicated by alternating foot processes of different electron densities); M, mitochondria; N, nephron; N(EC), nucleus of endothelial cell; N(P), nucleus of podocyte cell; P, podocytes; RBC, red blood cell; US, urinary space. Scale bars: 1 μm for C; 500 nm for D and E. N=3 for both.

Loss of lmx1ba delays cartilage growth and maturation. (A) Confocal images of z-stack projections of the lower jaw at 3 dpf and 5 dpf in wild type and lmx1ba mutants immunostained for Col2a1 (white). Scale bars: 50 μm. Schematics indicating where measurements were made for (B) length of lower jaw, (C) length of Meckel's cartilage and (D) width of Meckel's cartilage (left), with quantification of results (right). N=7 for wild type; 8 for lmx1ba mutants. Student's unpaired t-test performed for all where ****P<0.0001. Confocal z-stack projections of the (E) lower jaw and (F) lower jaw joint at 5 dpf in wild type and lmx1ba mutants, immunostained for EdU (cyan) and Collagen Type II (Col2a1; red) following 24-h treatment with EdU Click-iT. Scale bars: 50 μm for E and 10 μm for F. Number of EdU-Col2a1-positive cells quantified for (G) the whole lower jaw and (H) jaw joint. N=6 for wild type; 8 for lmx1ba mutants. Student's unpaired t-test performed for all where *P=0.0362. (I-L) Electron microscopy of ethmoid plate in wild type and lmx1ba mutants at 5 dpf. (I,K) Pink arrowheads highlight areas of non-uniformity and non-intercalating chondrocytes. Scale bars: 5 µm. (M) Number of chondrocytes on periphery of cartilage that are not aligning down central line of stack. Calculated as percentage of total cell number along ethmoid plate in one section per fish. (N) Number of vesicles present per chondrocyte. Mann–Whitney unpaired t-test performed where, ****P=0.0001. N=30 chondrocytes total from three larvae per genotype.

Loss of Lmx1b affects muscle formation in the trunk but not in the head. Confocal z-stack projections of wt and dKO mutants at 5 dpf immunostained with myosin, showing the whole body, scale bars: 250 μm, (A) and zoomed in regions of trunk muscle where defects in muscle formation increase in severity from no defect in wild type (B) to increasing number of defects in dKOs (C-E). Scale bars: 50 μm. (C,D) Pink asterisks indicate large gaps between myofibrils; (D) white arrowheads indicate abnormal fibre branching. (F) Percentage of larvae showing muscle defects at different severities quantified for wt and dKO larvae at 5 dpf. Differences in muscle formation between wt and dKO larvae at 3 dpf and 5 dpf quantified as (G) somite area and (H) the percentage coverage of muscle (myosin) staining per somite across all genotypes. Each data point is an average of measurements taken from three separate somites per fish. Schematics (right) show how measurements were made. N=11 for wild type and dKO mutants. Welch's t-test performed for G where ***P=.0001. Student's unpaired t-tests performed for H where *P=.01 and **P=.0092. (I) Confocal maximum projections of wt and dKO mutant larvae at 5 dpf immunostained for muscle (green) and co-stained with DAPI (blue). Scale bars: 100 μM. (J) Confocal z-stack projections of the lower jaw of wild type and dKO mutants immunostained for the tendon marker Thbs4 (cyan) and Collagen type II (Col2a1; red) at 5 dpf. Scale bars: 50 μM. (K) Schematic showing location and names of key muscles in the lower jaw of zebrafish. Muscles measured here are filled in in green. (L) Graphs showing quantification of muscle fibre length and width in the intermandibularis posterior (imp) and interhyal (ih) muscles at 5 dpf. N=5 for wild type and 7 for dKOs. Two datapoints plotted per fish, per age as measurements taken from right and left side of the lower jaw. Student's unpaired t-test performed at each age for all graphs.

Lmx1b is required for notochord cell inflation and body elongation. (A) Confocal images of the notochord in wild type and dKO larvae immunostained for Col2a1 at 3 dpf to show notochord sheath and aid measurement of notochord width. Pink line shows where notochord width was measured from. Scale bar: 50 μm. (B) Graph showing width of notochord as a percentage of body width at 3 dpf and 5 dpf. N=7 for wild type at 3 dpf and 5 at 5 dpf; 8 for dKO at 3 dpf and 7 at 5 dpf. Student's unpaired t-test performed at each age, where *** P=0.0009 and *** P=0.0004, respectively. (C) Confocal images of BODIPY stained larvae at 3 dpf. Scale bar: 50 μm. Low power electron microscopy images of notochord in (D) wild type and (E) dKO larvae at 3 dpf. Scale bar: 25 μm. Insets (Di, Eii) show false colouring for vacuolated cells (pink); unvacuolated cells (blue); nearby sheath cells (green); notochord sheath (yellow) and intracellular contents of vacuolated cells (purple).