Serrano et al., 2019 - Inhibition of Notch signaling rescues cardiovascular development in Kabuki Syndrome. PLoS Biology   17:e3000087 Full text @ PLoS Biol.

Fig 1 Generation and transcriptome profiling of <italic>kmt2d</italic> zebrafish mutants.

(A) Schematic of zebrafish kmt2d gene showing the 19 bp deletion allele (zy59) and its predicted amino acid change in the protein. The gRNA was designed to target exon 8 (red shaded exon) at the 5ʹ end of the gene. The resulting 19 bp deletion is predicted to cause an early stop codon at the level of the PHD tandem domains (blue) located at the N terminus of the protein. The anti-Kmt2d antibody epitope is located after the PHD domains and before the HMG domain (green), allowing the validation of the early stop in kmt2dzy59 zebrafish null mutant. (B) Lateral views of zebrafish wild-type sibling embryos (a–c) and kmt2d mutants for 3 different alleles: kmt2dzy59 (a, d, g), kmt2dzy58 (b, e, h), and kmt2dzy60 (c, f, i) at 3 dpf. All 3 alleles shared the same phenotypic characteristics: microcephaly (arrowhead), heart edema (arrow), and mild to moderate body axis defects. Variable expressivity was observed in all the analyzed mutant alleles (Classes I to III). Scale bar = 500 μm. (C) Variable expressivity was analyzed in 3 different embryo clutches resulting from a heterozygous by heterozygous cross for each mutant allele. Embryos were ranked in 3 different classes based on the severity of the phenotype. Class I, wild-type and heterozygous siblings with no phenotype; class II, mutants with microcephaly and heart edema; class III, mutants with microcephaly, heart edema, and shorten body axes. Percentages of different clutches were calculated per the total number of living embryos for each genotype. Chi-square test (p = 0.14) and binomial test (p = 0.09) were performed to assess mendelian ratios considering heterozygous embryos within the Class I category. There was no significant discrepancy between obtained and expected percentages of embryo phenotype. Values for each data point can be found in S1 Data. (D) MA plot of differentially expressed genes from RNA-seq of individual kmt2dzy59 mutants (n = 6) versus wild-type (n = 6) sibling embryos at 1 dpf. The log10 of mean normalized counts are plotted against the log2 fold changes for each gene tested. Green horizontal lines represent 2-fold change differences. Negative log2 fold changes represent genes with reduced expression in the mutants relative to wild type. Both light pink points (no outline) and color-coded points (with outline) represent significantly differentially expressed genes (FDR adjusted p < 0.05). The top 50 genes (ranked by FDR adjusted p-value) were classified into 6 categories based on expression data, bibliography, phenotype information, and gene ontology (S1 Table). Raw data used for this analysis can be found in the following repository: https://b2b.hci.utah.edu/gnomex/gnomexFlex.jsp?requestNumber=475R. (E) GSEA. Analysis was performed by converting zebrafish gene names to human gene names using genes with a one-to-one ortholog relationship retrieved from Ensembl Compara database. The number of resulting genes identifiers analyzed was 9,128 out of 33,737 (Genome build Zv9, Ensembl annotation released version 79). Adjusted p-values were calculated per category. NES of gene sets with a FDR of 5% are plotted to summarized GSEA results. dpf, days post fertilization; ECM, extracellular matrix; FDR,; GO, gene ontology; GSEA, gene set enrichment analysis; HMG, High Mobility Group domain; Kmt2d, Histone-Lysine N-Methyltransferase 2D; NES, Normalized Enrichment Score; PHD, Plant Homeo Domain; RNA-seq, RNA sequencing; SET: Su-Enhancer-of-zeste and Trithorax domain.

Fig 2 <italic>kmt2d</italic> mutants exhibit anomalous palate development and middle ear structures defects.

(A-D) Alcian Blue/Alizarin Red staining of cartilage and bone showing zebrafish homologous structures for mammalian palate (neurocranium) and middle ear (jaw structures). Ventral view of zebrafish sibling control (A) and kmt2dzy59 mutant (C) embryos at 4 dpf with corresponding simplified cartoons (B, D). kmt2dzy59 mutants had severe hypoplasia of visceral cartilages (C; dark blue, D) and neurocranium (C; light blue, D) when compared with wild-type siblings (A, B). The ethmoid plate was present but displayed abnormal development (eth). The cartilages that pattern the jaw in the mandibular (m, pq) and hyoid (ch, hm) arches were absent (m*) or drastically reduced. Pharyngeal arches were absent (cb1-5). The specific structures that are considered mammalian homologs of palate and middle ear are highlighted in orange (eth, me). (E) Quantification of width/length ratio of the ethmoid plate in wild-type siblings and kmt2dzy59 mutants. In all mutant embryos analyzed, the ethmoid plate was present but had a significantly reduced length/width ratio when compared with wild-type siblings. Statistical analysis was carried out using two-tailed t test, p < 0.0001, n = 10 per genotype. Values for each data point can be found in S1 Data. (F) Embryos were classified according to the degree of development of homolog structures for mammalian middle ear. The categories were present, mispatterned, or absent. Qualitative assessment was plotted for percentage of embryos per genotype (n = 10 per genotype). Values for each data point can be found in S1 Data. bh, basihyal; cb, ceratobranchial; ch, ceratohial; dpf, days post fertilization; eth: ethmoid plate; hm, hyomandibula; ie, inner ear; m, Meckel’s; me, middle ear mammalian homologs; ot, otolith; pq, palatoquadrate.

Fig 3 <italic>kmt2d</italic> mutants exhibit hypoplastic heart and aberrant endocardial cells morphology.

(A) Confocal images of whole mount immunofluorescence for wild-type zebrafish Kmt2d protein expression at 2 dpf (ventral view). Kdrl (endothelium and endocardium) and MF20 (myosin) were use as context marker for cardiovascular tissues and myocardium. Kmt2d expression was found in the nuclei of myocardial (Aʹ, asterisk in example cells) and endocardial (A", arrows) cells in zebrafish heart (Aʹ and A" are zoomed images in ventricle area). Scale bar = 100 μm. (B) Cartoon showing Kmt2d nuclear expression in both myocardium and endocardium tissue of zebrafish heart. Arrows in Aʹ, A", and F are showing Kmt2d endocardial expression in the same set of cells. (C–D) Confocal images of wild-type tg(kdrl:GFP) sibling (C, Cʹ) and kmt2dzy59;tg(kdrl:GFP) (D, Dʹ) embryos at 5 dpf. Ventral view. kmt2dzy59 mutant had aberrant cardiac morphology with significantly reduced ventricle size (D). Myocardial cell labeled with Alcama antibody (C–D, red) showed normal cell morphology in both genotypes. Maximum intensity projections of GFP channel evidenced abnormal morphology of endocardial cells (Dʹ). Some endocardial cells exhibited cell protrusions (Dʹ, arrowheads). (E, F) High magnification images of the heart ventricle chamber in a middle-plane view from the three-dimentional data set in wild-type sibling embryo (E, higher magnification of C) and kmt2dzy59 mutant (F, higher magnification of D) at 5 dpf. Z-stack analysis of the data set revealed that endocardial cells from the ventricle are organized in concentric layers (F, arrows) filling up the cardiac lumen in kmt2dzy59;tg(kdrl:GFP) mutants (E, F, asterisk). (G) Ventricle cavity volume measurements in 5 embryos per genotype. Statistical analysis was carried out using two-tailed t test, p < 0.0001. Scale bar = 25 μm. Values for each data point can be found in S1 Data. (H, I) Ventricle myocardial (H) and endocardial (I) cells quantification at 3 and 4 dpf in zebrafish wild-type siblings and kmt2dzy59 mutants. Embryos were processed for immunofluorescence against myosine heavy chain (MF20) and GFP (Kdrl). Nuclei were stained with DAPI for cell quantification with Imaris software. (H) Myocardial cells, t tests per time point, p-values as follow: 3 dpf p = 0.129, effect size = 24.6. 4 dpf p = 0.906, effect size = −2. Bonferroni corrected p-values for 2 t tests p adjusted values, 3 dpf = 0.258 and 4 dpf = 1.000. (I) Endocardial cells, t tests per time point, p-values as follow: 3 dpf p- = 0.0007, effect size = 29. 4 dpf p = 0.4139, effect size = 14.6. Bonferroni corrected p-values for 2 t tests p adjusted values, 3 dpf = 0.0015 (value reported in figure) and 4 dpf = 0.8278. Values for each data point can be found in S1 Data. avc, atrio-ventricular canal; dpf, days post fertilization; kdrl, kinase insert domain receptor like; kmt2d, Histone-lysine N-methyltransderase 2D; oft, outflow tract; ventr, ventricle.

Fig 4 <italic>kmt2d</italic><sup><italic>zy59</italic></sup> mutants fail to develop AAs and exhibit misspatterned cephalic vasculature.

(A-B) Ventral view of vasculature in wild-type tg(kdrl:GFP) sibling (A) and kmt2dzy59;tg(kdrl:GFP) mutants at 5 dpf. Cephalic vascular architecture in kmt2dzy59 embryos was abnormal with reduced elongation in the anteroposterior body axis, loss of bilateral symmetry, and reduced HA (A, B). In kmt2dzy59 mutants, AA1 (A, B) was shorter with minimum elongation towards the anterior area of the embryo, the AA3 (A, B) was rudimentary,whereas AA4−6 were reduced or absent (A, B). Mouth and eyes in kmt2dzy59 showed primitive characteristics with differences in the thickness of the OV in particular (A–B). Note the abnormal endocardial component of the heart (A, B; he). Aʹ and B are simplified cartoons of the main differences in the AAs development in wild-type (Aʹ) and mutant (Bʹ) backgrounds. Branchial vasculature loops were removed to allow better visualization of aortic arch points of origin. (C-D) Lateral view of cephalic vasculature in wild-type tg(kdrl:GFP) sibling and kmt2dzy59;tg(kdrl:GFP) mutants at 7 dpf. kmt2dzy59 exhibited a complete absence of the vascular loops associated with AAs 3–6 with only a rudiment of the AA3 present (D, AA*). The general cranial vascular network was mispatterned with a particular strong impact in the CtA, ACeV, and VA (D; in orange). In kmt2dzy59 mutant, most vessels had reduced lumens with the exception of the NCA (C, D), DVC (C, F) and IOC (C, F) that show thicker vessel diameter. Note the reduced HA from a lateral view (D, HA in orange; scale bars = 100 μm). (E) Cell cycle profile analysis by FACS for 7 individual embryos per genotype at 7 dpf. The gates set up for nuclear staining (DAPI) in kdrl positive cells (endothelial and endocardial cells) are shown. (F) Cell cycle profiles for Kdrl positive cells in kmt2dzy59 mutants showed no significant difference in the percentage of G2/M cells at 3 dpf. In contrast, at 7 dpf, there was a significant decrease in number of dividing cells in kmt2dzy59 mutants when compared with wild-type siblings. Unpaired two-tailed t test, p = 0.407 n.s. for 3 dpf; p = 0.0002 for 7 dpf, n = 7 per genotype. Values for each data point can be found in S1 Data. AA, aortic arch; AA1, mandibular arch; AA2, hyoid arch; AA3, first branchial arch; AA4, second branchial arch; AA5 third branchial arch; AA6, fourth branchial arch; ACeV, anterior cerebral vein; CtA, central artery; DCV, dorsal ciliary vein; DLV, dorsal longitudinal vein; dpf, days post fertilization; ey, eye; FACS, fluorescent activated cell sorting; GFP, green fluorescent protein; HA, hypobranchial artery; kdrl, kinase insert domain receptor like; IOC, inner optic circle; MsV, mesencephalic vein; NCA, nasal ciliary artery; ORA, opercular artery; OV, optic vein; VA, ventral aorta.

Fig 5 <italic>kmt2d</italic><sup><italic>zy59</italic></sup> mutants fail to develop AA3 to AA6.

(A–B) Still images (MIP) from time-lapse live imaging performed from 2 dpf to 3 dpf. Cranial-lateral view at the level of AA development from wild-type sibling (A–A‴, S3 Video) and kmt2dzy59 mutant (B–B‴, S4 Video). Images were selected at the 0, 4, and 11 hour time points. Images were converted to grayscale and inverted for better visualization (A–A", B–B"). A‴ and B‴ show A" and B" without grayscale processing. Red asterisk (B") denotes abnormal vascular development of AA sprouts at the level of the ventral border of LDA. (scale bars = 50 μm). (C-D) Schematic cartoons highlighting the development of AA in wild-type embryos (C, Cʹ) and kmt2dzy59 mutants (D, Dʹ). C corresponds to A, Cʹ correspond to A", D corresponds to B and Dʹ to B". Red asterisk indicates ectopic AA sprouting. (E–F) Cranial-lateral view of vasculature in wild-type tg(kdrl:GFP) sibling (A) and kmt2dzy59;tg(kdrl:GFP) mutants at 3 dpf. After time-lapse experiment (A, B), embryos were released from agarose and processed for IF and confocal imaging. Wild-type embryos showed correct patterning and secondary sprouting of AA3 to AA6 (E). At 3 dpf, kmt2dzy59 mutants had abnormal development of primary sprouts of AA3 to AA6, which are thinner and atrophic. In contrast, AA1 and ORA were thicker and exhibit abnormal morphogenesis and endothelial cell protrusions (F, white arrowhead; scale bars = 25 μm). (G-I) Still images (MIP) from time-lapse live imaging performed from 3 dpf to 4 dpf. Cranial-lateral view at the level of AA development from wild-type sibling and kmt2dzy59 mutant. Images were selected at the 0, 10, 11.25, and 15 hour time points. Asterisks (G–G‴) denote normal vascular development in the area between AA1 and ORA. No blood vessel forms in this area in wild-type background. Arrowheads (H–H‴, I–I‴) indicate endothelial cells extending multiple filopodia and forming a new and ectopic blood vessel. White dashed rectangle (H) specifies zoomed area in (I). Images were set on grayscale and the look-up table was inverted for better contrast of tip cell-like endothelial cells in kmt2dzy59 mutant (I–I‴; scale bars = 100 μm). AA, aortic arch; AA1, mandibular arch; AA2, hyoid arch; AA3, first branchial arch; AA4, second branchial arch; AA5, third branchial arch; AA6, fourth branchial arch; dpf, days post fertilization; he, heart; hpf, hours post fertilization; kdrl, kinase insert domain receptor like; LDA, lateral ventral aorta; MIP, maximum intensity projection; ORA, opercular artery.

Fig 6 Notch signaling is hyperactive in <italic>kmt2d</italic> mutant endothelium/endocardium.

(A-B) Confocal images of the heart of wild-type noninjected control tg(tp1:EGFP)um14 embryo (A) and F0 kmt2d mosaic mutants tg(tp1:EGFP)um14 embryo injected with CRISPR/Cas9 targeting kmt2d (B). F0 kmt2d mosaic mutants injected embryos showed hypoplastic heart as seen in kmt2dzy59 null mutants (B). Notch positive cells (green) were mostly distributed in the atrio-ventricular valve of a 3 dpf embryonic heart. Some endocardial cells in the ventricle and outflow tract were also observed (A). In F0 kmt2d mosaic mutants, hearts showed a significant increase in the number of Notch positive cells in both ventricle and atrium. Ventral view of the heart, only middle sections of the whole data set are shown. Scale bar = 25 μm. (C) Quantification of the amount of Notch positive cells in ventricle and atrium of control and injected embryos. N = 7 per group, unpaired two-tailed t test, p = 0.0001 in ventricle t = 4.95, dF = 36, atrium p = 0.0001, t = 10.99, dF = 36. Values for each data point can be found in S1 Data. (D) RT-qPCR analysis of wild-type sibling control embryos and kmt2dzy59 mutants for some components of the Notch signaling pathway. The Notch transcription factor rbpja was significantly up-regulated kmt2dzy59 embryos, corroborating the results obtained in the F0 kmt2d mosaic mutant analysis. There were no significant differences found for notch1b and hes1. N = 4 per genotype with 2 technical replicates per gene and per genotype assessed; elfα was use as control gene. Ct values were normalized using α-tubulin as gene of reference; fold change of relative expression was calculated using the ΔΔCt method. Multiple t test p < 0.0001 for rbpja, t = 6.04, dF = 24. Values for each data point can be found in S1 Data. (E–F) Confocal images of wild-type sibling control embryos (E) and kmt2dzy59 mutants (F) showing Rbpj protein expression levels and patterning at 5 dpf. Ventral view of the heart, only MIP of half data set is shown. Scale bar = 25 μm. (G) Summarized data and statistics from FACS experiment performed in 14 individual samples (7 wild-type siblings and 7 kmt2d mutants). Wild-type tg(kdrl:GFP) siblings (A) and kmt2dzy59;tg(kdrl:GFP) mutants were collected at 3 dpf, were processed for IF, were digested, and were prepared for FACS. Unpaired two-tailed t test, % of EC cells Rbpj positives p = 0.63, n.s. t = 0.49, dF = 12; Rbpj MFI p = 0,0009, t = 4.35, df = 12. Values for each data point can be found in S1 Data. Altogether our results show that Notch pathway is hyperactivated in endocardial cells of kmt2dzy59 mutants and demonstrate that this increased Notch activity is consequence of up-regulated Notch pathway transcription factor Rbpj specifically in EC cells. To our knowledge, these results provide the first evidence of a regulatory link between Kmt2d and Notch signaling during developmental processes in vertebrates. Ct, cycle threshold; ΔΔCt, delta-delta cycle thrshold; dpf, days post fertilization; EC, endocardial cells; F0, filial 0; FACS, fluorescent activated cell sorting; G0, generation 0; IF, immunofluorescense; MIP, maximum intensity projection; oft, out flow tract; RT-qPCR, reverse transcription-quantitative polymerase chain reaction; ven, ventricle.

Fig 7 Inhibition of Notch pathway with DAPT rescues cardiovascular phenotypes in <italic>kmt2d</italic> mutants by enhancing cell proliferation in endothelial and endocardial cells.

(A-D) Confocal images of wild-type sibling (A, B) and kmt2dzy59 mutant (C, D) embryos at 5 dpf. DMSO as solvent control (A, C) and DAPT for Notch signaling inhibition (B, D) were applied to embryos of indicated genotypes. DAPT treatment affected cardiovascular development in wild-type embryos (B) as a consequence of Notch signaling inhibition. kmt2dzy59 mutants that were treated with DAPT had rescued AA development and partially rescued heart development (D) when compared with the kmt2dzy59 DMSO control group (C). Arrowheads indicate normal (A), disrupted (B, C), and rescued (D) AA. Note that cardiovascular phenotype was rescued in DAPT-treated mutants despite cardiac edema, as evidenced by the sternohyoideus deformation (asterisk; B, C, D, MF20 in magenta). (E) Schematic of Notch signaling pathway showing DAPT inhibition of γ-secretase activity, the second cleavage step of Notch receptor processing. DAPT thus prevents NICD release to the cytoplasm for nuclear import. (F) Cardiac ventricle volume measurements for DMSO control groups (wild type and kmt2dzy59) and DAPT treatment groups (wild type and kmt2dzy59). The volume of the ventricle chamber was significantly rescued in kmt2dzy59 mutant embryos after Notch pathway inhibition with DAPT. Embryos were randomly selected from 3 different clutches; measurements were blind to embryo genotypes, which were assayed by HRMA after measurement. Two-way ANOVA multiple comparison test adjusted p-values per each condition as follow: wild-type DMSO versus wild-type DAPT, p < 0.0001; wild-type DMSO versus kmt2dzy59 DMSO, p < 0.0001; wild-type DMSO versus kmt2dzy59 DAPT, p = 0.0159; wild-type DAPT versus kmt2dzy59 DMSO, p = 0.5482; wild-type DAPT versus kmt2dzy59 DAPT, p < 0.0001; kmt2dzy59 DMSO versus kmt2dzy59 DAPT, p = 0.0001. Values for each data point can be found in S1 Data. (G) Number of endothelial cells proliferating in the AA region in each group treatment. There were significantly more proliferating endothelial cells in the AA region of kmt2dzy59 mutant after DAPT treatment, indicating that the phenotype was rescued by increasing cell proliferation. Two-way ANOVA multiple comparison test adjusted p-values per each condition as follow: wild-type DMSO versus wild-type DAPT, p = 0.0001; wild-type DMSO versus kmt2dzy59 DMSO, p = 0.0001; wild-type DMSO versus kmt2dzy59 DAPT, p = 0.0006; wild-type DAPT versus kmt2dzy59 DMSO, p = 0.6043; wild-type DAPT versus kmt2dzy59 DAPT, p = 0.0091; kmt2dzy59 DMSO versus kmt2dzy59 DAPT, p = 0.0047. Values for each data point can be found in S1 Data. (H) Number of endocardial cells proliferating in the heart per experimental group. There were significantly more proliferating endocardial cells in kmt2dzy59 mutant hearts after DAPT treatment, indicating that the phenotype was rescued by increasing endocardial cell proliferation. Two-way ANOVA multiple comparison test adjusted p-values per each condition as follow: wild-type DMSO versus wild-type DAPT, p = 0.0051; wild-type DMSO versus kmt2dzy59 DMSO, p = 0.0307; wild-type DMSO versus kmt2dzy59 DAPT, p = 0.0013; wild-type DAPT versus kmt2dzy59 DMSO, p = 0.8106; wild-type DAPT versus kmt2dzy59 DAPT, p < 0.0001; kmt2dzy59 DMSO versus kmt2dzy59 DAPT, p < 0.0001. Cell count was blind and embryos were randomly selected from 2 different clutches (E, F). Genotype was confirmed after measurement by HRMA. Values for each data point can be found in S1 Data. AA, aortal arch; ADAM, containing a disintegrin and metalloprotease; dpf, days post fertilization; ER, endoplasmic reticulum; HRMA, High Resolution Melt Analysis; kdrl, kinase insert domain receptor like; MF20, Myosin Heavy Chain Antibody; NICD, Notch intracellular domain.

Fig 8 Pharmacological inhibition of Notch signaling reestablishes normal Rbpj protein levels in kmt2d mutant endothelial and endocardial cells.

(A–D) Confocal images of wild-type sibling (A, B) and kmt2dzy59 mutant (C, D) embryos at 3 dpf. Cephalic ventral views. DMSO as solvent control (A, C) and DAPT for Notch signaling inhibition (B, D) were applied to embryos of indicated genotypes. Rbpj protein levels (Red channel) are higher in DMSO-kmt2dzy59 mutants (C, Cʹ). Arrow in Aʹ–Dʹ is indicating endocardial Rbpj expression—note higher endocardial Rbpj levels in Cʹ. DAPT treatment restored Rbpj signal levels in kmt2dzy59 embryos (D, Dʹ) as a consequence of Notch signaling inhibition. Dashed line in Aʹ–Dʹ indicates heart outline. Red, Rbpj; green, GFP (to enhance endocardial/endothelial Kdrl:GFP transgenic label). (E-F) Summarized data and statistics from FACS experiments performed in 16 individual samples per time point (4 DMSO wild-type siblings and 4 DMSO kmt2d mutants; 4 DAPT wild-type siblings and 4 DAPT kmt2d mutants per time point). Wild-type tg(kdrl:GFP) siblings (A) and kmt2dzy59;tg(kdrl:GFP) mutants from each treatment (DMSO or DAPT) were collected at 2 dpf (E) and 3 dpf (F), processed for IF, digested, and prepared for FACS. Two-way ANOVA multiple comparison test adjusted p-values per each condition at 2 dpf as follow: DMSO:wild-type versus DMSO:kmt2dzy59: p = 0.7762, DMSO:wild type versus DAPT:wild type: p = 0.9179, DMSO:wild type versus DAPT:kmt2dzy59: p = 0.9849, DMSO:kmt2dzy59versus DAPT:wild type: p = 0.9823, DMSO:kmt2dzy59versus DAPT:kmt2dzy59: p = 0.9186, DAPT:wild type versus DAPT:kmt2dzy59: p = 0.9910. Two-way ANOVA multiple comparison test adjusted p-values per each condition at 3 dpf as follow: DMSO:wild type versus DMSO:kmt2dzy59: p = 0.0221 (significant *), DMSO:wild type versus DAPT:wild type: p = 0.5394, DMSO:wild type versus DAPT:kmt2dzy59: p = 0.9263 (no significant difference, DAPT rescue was to similar control wild type values), DMSO:kmt2dzy59versus DAPT:wild type: p = 0.0012, DMSO:kmt2dzy59versus DAPT:kmt2dzy59: p = 0.0039 (significant *, DAPT rescues kmt2d mutant Rbpj levels), DAPT:wild type versus DAPT:kmt2dzy59: p = 0.9839. ANOVA summary results: Interaction F (1, 12) = 4.173, p = 0.0637; Treatment F (1, 12) = 19.37, p = 0.0009, Genotype F (1, 12) = 9.202, p = 0.0104. Values for each data point can be found in S1 Data. dpf, days post fertilization; FACS, fluorescent activated cell sorting; GFP, green fluorescent protein; kdrl, kinase insert domain receptor like.

Fig 9 Pharmacological inhibition of Notch signaling rescues early endothelial phenotypes and suppresses ectopic blood vessel formation in zebrafish <italic>kmt2d</italic> mutants.

(A–H) Still images (MIP) from time-lapse live imaging performed from 2 dpf to 2.5 dpf (A–D) and 3 dpf to 3.5 dpf (E–H). wild-type tg(Kdrl:GFP) and kmt2dzy59;tg(Kdrl:GFP) samples were treated with DMSO (n = 12; A, C, E, and G) or DAPT (n = 12; B, D, F, and H) from 1 dpf to 2 dpf. After treatment, samples were washed and prepared for in vivo time-lapse imaging. Cranial-lateral view at the level of AA development corresponding to video as follow: A, S7 Video; B, S9 Video; C, S8 Video; D, S10 Video; E, S11 Video; F, S13 Video; G, S12 Video; H, S14 Video. Images were selected at last time points recorded. Videos and images were converted to grayscale and inverted for better visualization. Asterisk (B, C, and G = AA*) denotes abnormal or missing vascular development of AA sprouts at the level of the ventral border of LDA. (scale bars = 50 μm). AA, aortic arch; AA1, mandibular arch; AA2, hyoid arch; AA3, first branchial arch; AA4, second branchial arch; AA5 third branchial arch; AA6, fourth branchial arch; dpf, days post fertilization; hpf, hours post fertilization; kdrl, kinase insert domain receptor like; LDA, lateral ventral aorta; MIP, maximum intensity projection; ORA, opercular artery.

Fig 10 Model summary schematic.

Schematic cartoon highlighting cardiovascular defects in our genetic zebrafish model for Kabuki Syndrome. Notch signaling is identified as primary candidate pathway underlying the endothelial/endocardial phenotype. Pharmacological inhibition of Notch signaling was able to rescue the cardiovascular phenotype in KS mutants. AA, aortic arch; AV, atrioventricular.

Acknowledgments:
ZFIN wishes to thank the journal PLoS Biology for permission to reproduce figures from this article. Please note that this material may be protected by copyright. Full text @ PLoS Biol.