apc is required for the CNC contribution to craniofacial structures: (A) Wild-type (WT) and (B) apc^mcr/mcr larvae at ~30 somites (50 hpf). Arrowhead indicates the accumulation of amorphous tissue anterior to the heart and yolk mass. Scale bars in (A,B) are 200 µm. (C,D) Hematoxylin-eosin staining of WT (C) and apc^mcr/mcr (D) larvae sectioned at the level of the hindbrain. Hb: hindbrain; S: somite; N: notochord; E: esophagus; L: liver; PF: pectoral fin. Scale bars in (C,D) are 100 µm. (E) Fli1-GFP epifluorescence of WT (top two larvae) and apc^mcr/mcr larvae (lower two larvae; 30-somite stage). Arrowheads for WT larvae indicate anterior extent of GFP⁺ cells in the head, which is reduced or absent in mutants. (F,G) Confocal image of WT; Fli1-GFP (F) and apc^mcr/mcr; Fli1-GFP (G) larvae at 60 hpf, showing the vascular endothelium of trunk. (H,I) Confocal image of Fli1-GFP in WT (H) and apc^mcr/mcr (I) larvae at 60 hpf, showing CNC and vascular endothelium in branchial arches and head. mc: Meckel’s cartilage; pq: palatoquadrate cartilage; ch: ceratohyal cartilage. Numbers indicate gill arches 1–5.

Expression of early and late neural crest markers by in situ hybridization: (A,B) Similar sox10 expression at neural plate border in WT (A) and apc^mcr/mcr (B) embryos at 12 hpf. Genotypes in (A) (n = 30 WT or apc^mcr/+) and (B) (n = 7 apc^mcr/mcr) were defined by genomic PCR on individual larvae. (C,D) In WT embryos at 55 hpf (C), gsc is expressed in the mandibular (m, including Meckel’s cartilage), hyoid (h), and more posterior branchial arches, as well as two symmetric anterior domains, as described previously [44]. In apc^mcr/mcr larvae (D), gsc is undetectable in the mandibular and hyoid arches. (For WT, n = 15, and for apc^mcr/mcr, n = 7, based on morphological phenotype). All scale bars are 200 µm.

Transcriptomic analysis of apc loss of function larvae: (A) Heat map showing differential gene expression in WT versus apc^mcr/mcr larvae at 48 hpf. To the right is a list of a upregulated genes known to be induced by Wnt/β-catenin activation [45], including multiple Wnt-induced pathway antagonists. (B) RT-qPCR was used to validate changes in expression from RNA-seq data for the indicated mRNAs. GAPDH was used as an input control for RT-qPCR. Error bars show standard error. (C) GO analysis (Metascape [37]) shows the differential expression of genes associated with the indicated terms in apc^mcr mutants. (D) Heat map showing the differential expression of mRNAs encoding secreted molecules that regulate neural crest migration and axon guidance. In zebrafish, Complement c3 is encoded by “a” and “b” alleles and the c3a alleles were duplicated to six copies, c3a.1–c3a.6. The “c3a.x” designation for these duplicated genes should be distinguished from the more common designation of the polypeptide “C3a”, the extracellular chemoattractant derived via the proteolytic processing of C3 protein that mediates the co-attraction of CNC cells [11]. The heat map also shows the reduced expression of SDF1/CXCL12 and the differential expression of multiple semaphorins. (E) The abundance of c3a.1, c3a.2, c3a.3, and c3a.6 genes in WT and apc^mcr larvae at 48 hpf was based on mean normalized read counts from three independent replicates from RNA-seq data. c3a.5 was excluded because mean read counts were <100 and c3a.4 was not detected. Fold change in expression for apc^mcr/WT is shown below graph. Relative change in expression was confirmed for c3a.1 and c3a.6 by RT-qPCR (not shown). Error bars show standard error. ** indicates p < 0.01 (Student’s t-test). (F) GSEA analysis [41] for apc^mcr/WT zebrafish identifies parallels with the complement pathway and EMT, as well as c-Myc, canonical Wnt signaling, mTORC1 activation, and inflammatory signaling pathways (Supplemental Table S3).

Oncogenic mutations in apc cause widespread splice variations including mRNAs encoding regulators of cell migration: (A) The MAJIQ analysis of splicing variations, represented as changes in fractions spliced in (dPSI) for each splicing event filtered for dPSI ≥ 0.20 with a probability >0.90, identified 340 splicing variations in apc^mcr larvae. (B) The frequency of local splice variants, including exon skipping (ES), alternative 3′ splice site (A3SS), alternative 5′ splice site (A5SS), and complex splice variants. (C) The GO analysis of alternative spliced genes. (D) Voila output showing alternative splicing of sema3f (“a” allele). Red line indicates splicing from exon 5 to exon 6 and blue line indicates splicing from exon 5 to exon 7. Violin plots show the frequency of each splice form in WT (left) and apc^mcr mutants (right) at 48 hpf. (E). Diagram showing unspliced, nascent mRNA from exon 5–exon 8. Orange arrows indicate PCR primers for “long” splice form that includes exon 6; green arrows show the primer pair for constitutively spliced exons 7 and 8, used to measure all mRNA isoforms. Histogram shows relative abundance based on RT-qPCR for the “long” splice form containing exon 6 in WT and apc^mcr mutant larvae at 48 hpf. p < 0.0001 (Student’s paired t-test, n = 3 replicates).

Oncogenic mutations in apc cause widespread splice variations including mRNAs encoding regulators of cell migration: (A) The MAJIQ analysis of splicing variations, represented as changes in fractions spliced in (dPSI) for each splicing event filtered for dPSI ≥ 0.20 with a probability >0.90, identified 340 splicing variations in apc^mcr larvae. (B) The frequency of local splice variants, including exon skipping (ES), alternative 3′ splice site (A3SS), alternative 5′ splice site (A5SS), and complex splice variants. (C) The GO analysis of alternative spliced genes. (D) Voila output showing alternative splicing of sema3f (“a” allele). Red line indicates splicing from exon 5 to exon 6 and blue line indicates splicing from exon 5 to exon 7. Violin plots show the frequency of each splice form in WT (left) and apc^mcr mutants (right) at 48 hpf. (E). Diagram showing unspliced, nascent mRNA from exon 5–exon 8. Orange arrows indicate PCR primers for “long” splice form that includes exon 6; green arrows show the primer pair for constitutively spliced exons 7 and 8, used to measure all mRNA isoforms. Histogram shows relative abundance based on RT-qPCR for the “long” splice form containing exon 6 in WT and apc^mcr mutant larvae at 48 hpf. p < 0.0001 (Student’s paired t-test, n = 3 replicates).