Identifying in vivo genetic dependencies of melanocyte and melanoma development
- Department of Cancer Biology and Genetics, Memorial Sloan Kettering Cancer Center, United States
- Gerstner Sloan Kettering Graduate School of Biomedical Sciences, Memorial Sloan Kettering Cancer Center, United States
- Weill Cornell/Rockefeller/Sloan Kettering Tri-Institutional MD-PhD Program, Memorial Sloan Kettering Cancer Center, United States
- Cell and Developmental Biology Program, Weill Cornell Graduate School of Medical Sciences, United States
- Institute for Systems Genetics, NYU Grossman School of Medicine, United States
- Department of Cell Biology, NYU Grossman School of Medicine, United States
- Department of Genetics, Development and Cell Biology, Iowa State University, United States
- Department of Biomedical Engineering, NYU Tandon School of Engineering, United States
- Nuffield Department of Medicine, Ludwig Institute for Cancer Research, University of Oxford, United Kingdom
Figures
Figure 1 with 1 supplement
Targeted integration of Cas9 to the mitfa locus with GeneWeld knock-in system.
(A) Schematic depicting integration site and GeneWeld knock-in cassette. 5’ and 3’ homology arms are designed to target exon 2 of the zebrafish mitfa gene. The knock-in cassette includes p2A followed by Cas9 and BFP driven by the eye-specific promoter γ-crystallin. Created with BioRender.com. (B) Pipeline to produce F1 mitfaCas9 zebrafish. The GeneWeld knock-in vector and gRNAs targeting the mitfa genomic insertion site or specific sites on the knock-in vector (UgRNA) are co-injected into one-cell-stage wild-type (WT) zebrafish embryos. Embryos are screened for BFP+ eyes marking mosaic integration. These mosaic founder fish are then raised to adulthood and crossed with WT fish. The resulting embryos are screened for BFP+ eyes and sequenced to confirm precise integration. Created with BioRender.com. (C) Fluorescence in situ hybridization (FISH) chain reaction on 3 days post-fertilization (dpf) WT and mitfaCas9 embryos treated with 1-phenyl 2-thiourea (PTU). Arrows on the whole embryo image (WT 3 dpf embryo not treated with PTU) indicate the embryonic melanocyte stripe regions where mitfa and Cas9 expression is expected. The presence of mitfa and Cas9 RNA was assessed by confocal microscopy at 40x magnification. Maximum intensity projections are shown. n=17 WT embryos and n=16 mitfaCas9 embryos were screened. Scale bars, 100 µm.
Figure 1—figure supplement 1
Validation of Cas9 knock-in.
(A) Integration efficiency of mitfa-targeted P2A-Cas9, gamma-cry:BFP vector. n=4 independent experiments are shown with 50 embryos screened for each experiment. BFP+ eyes were detected in a total of 37/200 embryos. Error bars, SD. (B) Sanger sequencing results from four F1 fish with BFP+ eyes. Primers were designed with the forward primer located in the genomic mitfa locus and reverse primer within the Cas9 insertion cassette so that amplification can only occur if there is integration in the correct orientation. SnapGene was used to align sequences. (C) Images of 3-day-old and 3-month-old clutchmates from a mitfaCas9 in-cross. (D) Quantification of the percentage of mitfa-positive cells that have Cas9 expression in each embryo. The presence of mitfa and Cas9 RNA was assessed by confocal microscopy from fluorescence in situ hybridization (FISH) chain reaction on 3 days post-fertilization (dpf) wild-type and mitfaCas9 embryos. Out of 28 mitfa+ cells assessed across n=7 embryos, 24 had detectable Cas9 RNA.
Figure 2 with 1 supplement
Melanocyte lineage-specific knockout of albino using mitfaCas9 fish.
(A) Pipeline to generate F0 and F1 U6:gRNA; mitfa:GFP (MG-gRNA) zebrafish. Created with BioRender.com. (B) Adult mitfaCas9 MG-NT and MG-albino F0 fish. (C) Proportion of MG-NT (n=19) and MG-albino (n=25) F0 fish with loss of pigmentation phenotype. (D) Adult mitfaCas9 MG-NT and MG-albino F1 fish. (E) Pigmented area/mm2 calculated for n=5 fish/genotype within a defined rectangular region of interest (ROI) encompassing the top and middle melanocyte stripes. Two-sided Student’s t-test was used to assess statistical significance, ****p<0.0001; error bars, SD. (F) 3 days post-fertilization (dpf) mitfaCas9 MG-NT and MG-albino F2 fish. Representative embryos are shown for each genotype. (G) Mean gray value of head melanocytes calculated for n=10 embryos/genotype within a defined hexagonal ROI indicated as a red outline in F. Two-sided Student’s t-test was used to assess statistical significance, ****p<0.0001. (H) Schematic for CRISPR sequencing protocol. Created with BioRender.com. (I) CRISPR-seq results are shown for WT and n=3 independent F1 MG-albino fish. Only GFP-negative cells were isolated from WT fish. Results are shown as a fraction of sequences with indels calculated using CRISPResso.
Figure 2—figure supplement 1
albino knockout fish.
(A) Quantification of pigmentation for 1 month post-fertilization (mpf) mitfaCas9 MG-NT and MG-albino F2 fish. Pigmented area/mm2 calculated for n=12 fish/genotype within a defined rectangular region of interest (ROI) encompassing the top and middle melanocyte stripes. Two-sided Student’s t-test was used to assess statistical significance, ****p<0.0001; error bars, SD. (B) Color and GFP images of 1 mpf mitfaCas9 MG-NT and MG-albino F2 fish. (C) Pigmented area/mm2 calculated for n=12 fish/genotype within a defined rectangular ROI encompassing the top and middle melanocyte stripes. Two-sided Student’s t-test was used to assess statistical significance, ****p<0.0001; error bars, SD. (D) Color images of 3 mpf mitfaCas9 MG-NT and MG-albino F2 fish. (E) Indel chart for the albino locus produced using CRISPRVariants. **T insertion is observed across all samples, likely resulting from a PCR or sequencing artifact.
Figure 3
Melanocyte lineage-specific knockout of sox10 using mitfaCas9 fish.
(A) Adult mitfaCas9 MG-NT and MG-sox10 F0 fish. (B) Proportion of MG-NT (n=19) and MG-sox10 (n=16) F0 fish with disrupted stripes phenotype. (C) Adult mitfaCas9 MG-NT and MG-sox10 F1 fish. (D) The average width of melanocyte stripe 1D and xanthophore interstripe X0 was calculated for each fish by averaging 5 stripes/interstripe width measurements. N=5 fish per genotype. Two-sided Student’s t-test was used to assess statistical significance, ***p<0.001; error bars, SD. (E) F1 MG-NT (n=5) and MG-sox10 (n=5) adult fish were treated with epinephrine, and melanocytes were counted within a defined rectangular region of interest 3.46 mm × 2.54 mm encompassing the top and middle melanocyte stripes. Two-sided Student’s t-test was used to assess statistical significance, *p<0.05; error bars, SD. (F) Schematic of neocuproine (neo) experimental setup. Adult mitfaCas9 MG-NT (n=5) and MG-sox10 (n=5) F1 fish were treated with neocuproine for 24 hr to ablate melanocytes, then imaged at days 7, 15, and 70 to measure regeneration of melanocytes compared to day 0. Created with BioRender.com. (G) Quantification of melanocyte regeneration. Fish were treated with epinephrine prior to imaging to enable counting of melanocytes. Two-sided Student’s t-test was used to assess statistical significance, **p<0.01; error bars, SD. (H) Representative images are shown for MG-NT and MG-sox10 fish pre- and post-neocuproine treatment.
Figure 4 with 1 supplement
Non-autonomous function of tuba1a/tuba1c on melanocytes.
(A) 4 days post-fertilization (dpf) zebrafish embryos injected with either NT or tuba1a/c Alt-R CRISPR Cas9 gRNAs. (B) Percentage of NT or tuba1a/c Alt-R zebrafish embryos with dispersed melanocyte phenotypes. N=2 independent experiments. (C) Survival percentage is shown for NT and tuba1a/c Alt-R embryos. Embryos were counted at 24 hr post-fertilization (hpf) and again at 14 dpf to determine survival. (D) 4 dpf mitfaCas9 MG-tuba1a/c F2 embryos (BFP+ eyes) compared to sibling controls with no mitfaCas9 (BFP- eyes). Representative embryos are shown for each genotype. (E) Pigmented area/mm2 calculated for n=19 embryos/genotype from two independent MG-tuba1a/c F2 clutches. Two-sided Student’s t-test was used to assess statistical significance, ns: no significance. (F) 4 dpf Alt-R-injected zebrafish embryos imaged before and after epinephrine (epi) treatment. (G) Pigmented area/mm2 calculated for n=20 embryos/genotype from two independent clutches. Two-sided Student’s t-test was used to assess statistical significance, ****p<0.0001.
Figure 4—figure supplement 1
Knockout of tuba1a/c.
(A) 6 days post-fertilization (dpf) zebrafish embryos injected with either NT or tuba1a/c sg1 Alt-R CRISPR Cas9 gRNAs. (B) Validation of Alt-R CRISPR Cas9 tuba1a/c sg1 targeting with Sanger sequencing. Synthego Ice software estimated 88% indels in the tuba1a locus and 97% indels in the tuba1c locus. (C) Validation of Alt-R CRISPR Cas9 tuba1a-specific gRNA targeting with Sanger sequencing. Synthego Ice software estimated 97% indels in the tuba1a locus. (D) Validation of Alt-R CRISPR Cas9 tuba1c-specific gRNA targeting with Sanger sequencing. Synthego Ice software estimated 56% indels in the tuba1c locus. (E) Validation of Alt-R CRISPR Cas9 tuba1a/c sg2 targeting with Sanger sequencing. Synthego Ice software estimated 88% indels in the tuba1a locus and 86% indels in the tuba1c locus. (F) Percentage of NT or tuba1a/c sg2 Alt-R zebrafish embryos with dispersed melanocyte phenotypes. N=2 independent experiments. (G) Adult mitfaCas9 MG-tuba1a/c F0 fish. Image is representative of n=7 F0 fish. (H) Adult mitfaCas9 MG-tuba1a/c F1 fish. Image is representative of n=14 F1 fish. (I) CRISPR-seq results are shown for wild-type (WT) and n=2 F1 MG-tuba1a/c fish sorted for GFP+ cells. Results are shown as a fraction of sequences with indels calculated using CRISPResso.
Figure 5 with 1 supplement
Generation of a zebrafish melanoma model in mitfaCas9 fish.
(A) Adult mitfaCas9 MG-NT; mitfa:TdTomato;U6:NT (MTdt-NT) and MG-ptena; mitfa:TdTomato;U6:ptenb (MTdT-ptenb) F0 fish. (B) Proportion of MG-NT; MTdT-NT (n=14) and MG-ptena; MTdT-ptenb (n=14) F0 fish with disrupted stripes phenotype. (C) Schematic of zebrafish tumorigenesis assay. Indicated plasmids are injected into one-cell-stage embryos from crosses between mitfaCas9 and wild-type (WT) fish. The mitfa:BRAFV600E plasmid includes cardiac-specific cmlc2:GFP. Embryos were sorted for GFP+ hearts and BFP+ eyes, and fish were screened every 2 weeks for tumors. Created with BioRender.com. (D) Tumor-free survival curve. N=2 independent experiments. Tumors were tracked over the course of 30 weeks. Log-rank (Mantel-Cox) test was used to assess statistical significance, ***p<0.001; ****p<0.0001; ns: no significance. (E) Histology was performed on one fish from each indicated injection group. Dotted lines indicate the site of sectioning. Red chromogen was used for all immunohistochemical (IHC) staining. Scale bars, 50 µm. (F) CRISPR-seq results are shown for normal skin dissected from WT fish and tumors dissected from injection conditions 2, 3, and 4. Results are shown as a fraction of sequences with indels calculated using CRISPResso.
Figure 5—figure supplement 1
CRISPR-seq for tumor suppressor genes.
(A) Indel chart for the ptena locus produced using CRISPRVariants. (B) Indel chart for the ptenb locus produced using CRISPRVariants. (C) Indel chart for the p53 locus produced using CRISPRVariants.
Figure 6 with 1 supplement
Melanoma-specific knockout of sox10 reduces tumor burden and induces phenotypic switching.
(A) Schematic of zebrafish tumorigenesis assay. Indicated plasmids are injected into one-cell-stage embryos from crosses between mitfaCas9 and wild-type (WT) fish. Embryos were sorted for GFP+ hearts and BFP+ eyes, and fish were screened every 2 weeks for tumors. Created with BioRender.com. (B) Tumor-free survival curve. N=3 independent experiments. Tumors were tracked over the course of 50 weeks. Log-rank (Mantel-Cox) test was used to assess statistical significance, ****p<0.0001. (C) Histology is shown for one fish from each injection group. Dotted lines indicate the site of sectioning. Red chromogen was used for all immunohistochemical (IHC) staining. Scale bars, 50 µm. (D) Color thresholding was used on IHC images to calculate the percentage of nuclei that stained positive for sox9. NT-1 n=1222 cells, NT-2 n=1620 cells, NT-3 n=2185 cells, sox10-1 n=970 cells, sox10-2 n=594, sox10-3 n=1046. Cells from n=3 images were analyzed for each tumor. (E) Schematic of scRNA-seq experimental setup from Wouters et al. Patient-derived cell lines were treated with siRNAs targeting SOX10 or NTC, and scRNA-seq was conducted at 72 hr. (F) UMAP of scRNA-seq dataset for MM057 cell line from Wouters et al. Cells from siSOX10 and siNT conditions are labeled. (G) Normalized expression of Sox10 per cell in UMAP space. (H) Violin plots of normalized expression of Sox10 per cell. Median is shown as dashed red line. Wilcoxon rank sum test was used to assess statistical significance, ****p<0.0001. (I) Normalized expression of Sox9 per cell in UMAP space. (J) Violin plots of normalized expression of Sox9 per cell. Wilcoxon rank sum test was used to assess statistical significance, ****p<0.0001.
Figure 6—figure supplement 1
In vivo and in vitro targeting of Sox10 leads to upregulation of Sox9.
(A) Histology is shown for two additional fish from each injection group. Dotted lines indicate the site of sectioning. Red chromogen was used for all immunohistochemical (IHC) staining. Scale bars, 50 µm. (B) Sox10 intensity per nucleus from NT and sox10 KO tumors. Intensity was calculated from immunofluorescence staining of Sox10. Red lines indicate mean with 95% CI. Each point represents one nucleus. NT-1 n=1334 cells, NT-2 n=1606 cells, NT-3 n=1204 cells, sox10-1 n=869 cells, sox10-2 n=998, sox10-3 n=1046. Cells from n=3 images were analyzed for each tumor. (C) UMAP of scRNA-seq dataset for MM074 cell line from Wouters et al. Cells from siSOX10 and siNT conditions are labeled. (D) Normalized expression of SOX10 per cell in UMAP space in MM074 cell line. (E) Normalized expression of SOX9 per cell in UMAP space in MM074 cell line. (F) Violin plots of normalized expression of SOX10 per cell. Median is shown as a dashed red line. Wilcoxon rank sum test was used to assess statistical significance, ****p<0.0001. (G) Violin plots of normalized expression of SOX9 per cell. Wilcoxon rank sum test was used to assess statistical significance; ns: no significance. (H) UMAP of scRNA-seq dataset for MM087 cell line from Wouters et al. Cells from siSOX10 and siNT conditions are labeled. (I) Normalized expression of SOX10 per cell in UMAP space in MM087 cell line. (J) Normalized expression of SOX9 per cell in UMAP space in MM087 cell line. (K) Violin plots of normalized expression of SOX10 per cell. Median is shown as a dashed red line. Wilcoxon rank sum test was used to assess statistical significance, ****p<0.0001. (L) Violin plots of normalized expression of SOX9 per cell. Wilcoxon rank sum test was used to assess statistical significance, ****p<0.0001.
Additional files
-
MDAR checklist
- https://cdn.elifesciences.org/articles/100257/elife-100257-mdarchecklist1-v1.docx
-
Supplementary file 1
Sequences of primers, gRNAs, probes, and gBlock used in this study.
- https://cdn.elifesciences.org/articles/100257/elife-100257-supp1-v1.docx
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Identifying in vivo genetic dependencies of melanocyte and melanoma development
eLife 13:RP100257.
https://doi.org/10.7554/eLife.100257.3
