Data from: Buffering of genetic defects in animal development by regeneration programs
Data files
Aug 24, 2026 version files 11.39 GB
-
Archive.zip
11.39 GB
-
README.md
26.11 KB
Abstract
Regeneration programs enable animals to restore damaged or lost tissues, and the range of stimuli for these programs is incompletely understood. Here, we used zebrafish, a vertebrate species with exceptional regenerative capacity, to identify chemically induced mutations that alter regeneration-associated gene activation. Transgenic zebrafish with a permissive promoter and EGFP cassette inserted in the vicinity of the pro-regenerative factor gene fgf20a were mutagenized, and larvae homozygous for ENU-induced mutations were assessed for disruptions in fgf20a-directed reporter gene expression following fin fold amputation. One line was identified with heritable, elevated fgf20a:EGFP presence in the absence of experimental injury, localized to regions of fin fold tissue undergoing degeneration. Whole-genome sequencing (WGS) identified a mutation within exon 72 of the fraser syndrome 1 (fras1) gene, mutated in patients with inherited skin disease. fras1 mutant larvae spontaneously displayed elevated expression of other known injury/regeneration-responsive reporter lines in fin fold, and zebrafish crispants for homologs of other genes mutated in human developmental diseases also displayed regeneration-associated gene expression in regions of dysmorphology. Tempering Fgf signaling by transgenic expression of a dominant-negative Fgf receptor in fras1 mutants exacerbated the disease phenotype. Our findings provide evidence that regeneration programs are harnessed in response to developmental defects caused by genetic mutations, potentially buffering deleterious phenotypes.
Dataset DOI: 10.5061/dryad.jdfn2z3sm
Description of the data and file structure
This dataset Archive.zip includes data collected to assemble the figures and figure supplements in Ando et al. "Buffering of genetic defects in animal development by regeneration programs". Data include raw and processed images from fluorescence microscopy; xlsx spreadsheets with measurements; prism files for statistical tests and graphs; qPCR data files; and genome sequencing analysis files.
Folder: Figure 1
This figure describes an ENU (N‑ethyl‑N‑nitrosourea) mutagenesis screen designed to identify mutations that alter regeneration-responsive expression of fgf20a, a pro-regenerative factor regulated by tissue regeneration enhancer elements (TREEs) and tissue regeneration silencer elements (TRSEs). The enhancer trap reporter line HGn21A (fgf20a:EGFP) uses the fgf20aE48 TREE to visualize regeneration-induced fgf20a expression through EGFP (enhanced green fluorescent protein) fluorescence. ENU-treated homozygous male reporter fish were crossed with homozygous reporter females to generate F1, F2, and F3 generations that were screened for mutations affecting injury-induced fluorescence. Fin fold amputation induced fgf20a:EGFP expression at the injury site, recapitulating the normal injury response of fgf20a. A heritable mutant called budlight exhibited a shortened body and elevated fgf20a:EGFP expression in the degenerating fin fold even in the absence of experimental injury. Quantification of fgf20a expression showed that levels increased in injured wild-type larvae relative to uninjured controls and were highest in budlight mutants, with statistical significance determined by a two-sample t-test assuming unequal variances and reported as P values. Analyses were performed in larvae at 4 dpf (days post-fertilization) and 1 dpa (day post-amputation), data are presented as mean ± SEM (standard error of the mean) from n = 3 pooled biological replicates containing 20 larval tails per replicate, and images include 100 μm (micrometer) scale bars.
Sub-folder: 1C
File: 1C (S1B) ENU screen result.xlsx
File: 1C (S1B) ENU screen result.csv
Description: original statistical data and descriptions of phenotypes for 1C (and S1). Values include family ID number and the ratio of animals with phenotypes.
Sub-folder: 1D
Sub-sub folder: Raw (czi files)
Description: original czi files
Files: exported tif files
Description: named by animal number and injury context. There are 3-4 different animals each for uninjured and regeneration conditions.
Sub-folder: 1E
Sub-sub folder: Raw (czi files)
Description: original czi files
Files: exported tif files
Description: named by animal number and genotype. There are 3-4 different animals each for WT siblings and budlight mutants (1E).
Sub-folder: 1F
File: 1F fgf20a qPCR, WT uninj, WT 1dpa, budlight at 4dpf.pzfx
Description: data sheet with Cq values for fgf20a qPCR in larval tails.
Folder: Figure 2
Whole-genome sequencing was used to identify the mutation responsible for the budlight phenotype by pooling homozygous mutant larvae with elevated fgf20a:EGFP expression at 5 dpf (days post-fertilization) for genomic DNA extraction, library preparation, Illumina sequencing, and variant analysis. The causal mutation was identified in fras1 as a nonsense mutation consisting of a T-to-A substitution in exon 72 that converts a tyrosine codon into a premature stop codon. Targeted CRISPR-Cas9 mutagenesis of fras1 in F0 embryos reproduced the budlight phenotype and induced ectopic activation of the fgf20a:EGFP reporter in degenerating fin fold and craniofacial tissues. In contrast, control embryos displayed only the normal developmental domains of fgf20a:EGFP expression. Similar ectopic reporter activation and developmental abnormalities were observed after CRISPR-Cas9 disruption of zebrafish orthologues of human developmental disease-associated genes, including pex1, tbx5a, grip1a, and frem2a. Weak and strong levels of reporter induction were identified in different tissues, with yellow arrows indicating weak activation and green arrows indicating strong activation of fgf20a:EGFP. Quantification of the CRISPR-Cas9 F0 screen classified larvae as having no detectable phenotype, minor defects with ectopic reporter expression, or major defects with ectopic reporter expression, and statistical significance was determined using Fisher’s exact test with multiple-testing correction, where p < 0.05 and p < 0.01 indicated significant effects.
Sub-folder: 2A, B
Description: WGS analysis uploaded to GEO folder is empty. Data have been uploaded to GEO as indicated in the manuscript and not included here.
Sub-folder: 2C
Description: jpeg images of the 3 representative embryos shown as examples in Figure 2C. Naming is genotype and location on figure panel.
Sub-folder: 2D-G
Description: jpeg images of the 1-2 representative embryos shown in each panel. Naming is genotype and location on figure panel.
Sub-folder: 2H
Sub-sub-folders: ASPA, BMP1A, COL1A1A, COL1A2, COL2A1, Control, EBP, FGFR3, FLNB, FRAS1, FREM2A, GRIP1A, LMNA, NEK1, PEX1, ST14, TBX5A
Description: images of larvae from experiments for control crispant perturbations as well as perturbation of each of the genes tested in the screen. Each subfolder contains raw CZI and jpeg images for larvae that were used for ratio quantification in 2H. Each larva is considered as one replicate/count. Naming of files includes genotype and animal number. Additionally, there is a spreadsheet file “HumanGENESscreen_QUANT” in .xlsx and .csv formats where phenotypes are quantified across all genes (values are numbers of animals showing the indicated phenotype), and Fisher’s test data provided.
Folder: Figure 3
This figure provides evidence that fras1 mutants activate a broad regeneration-associated genetic program using multiple tissue regeneration enhancer element (TREE) reporters, including REN (runx1-linked enhancer) and LEN (leptin b-linked enhancer) driving EGFP (enhanced green fluorescent protein) expression. Reporter analyses showed that fgf20a:EGFP, RENcfos:EGFP, and LENP2:EGFP are normally induced during fin regeneration, with fgf20a expressed in epithelia, mesenchyme, notochord, and muscle, while RENcfos and LENP2 are activated primarily in notochord and mesenchyme. The reporter constructs contained either a 100 base pair (bp) mouse cfos promoter or a 2 kilobase (kb) zebrafish leptin b promoter linked to EGFP. In uninjured fras1 mutants, ectopic reporter activation occurred in notochord and mesenchymal tissues for RENcfos:EGFP and in mesenchyme for LENP2:EGFP, indicating regeneration-like signaling in the absence of injury. Quantification of EGFP-positive areas in distal tail regions at 4 dpf (days post-fertilization) showed elevated reporter activity in mutants compared with uninjured wild-type larvae and levels comparable to or approaching those seen in regenerating wild-type larvae at 1 dpa (day post-amputation); data are presented as mean ± SEM (standard error of the mean) with n = 3–7 biological replicates and analyzed using two-way ANOVA (analysis of variance). PCA (principal component analysis) of bulk RNA-seq (RNA sequencing)datasets from uninjured wild-type larvae, regenerating tails, and fras1 mutants revealed that the first two principal components (PC1 and PC2) explained 40.1% and 33.6% of total variance, respectively, and positioned mutants near regenerating samples in transcriptional space. Correlation and heatmap analyses further demonstrated that many genes altered during regeneration were similarly regulated in fras1 mutants, with concordant changes assessed using log2FC (log2 fold change) values, Pearson and Spearman correlation coefficients, row-scaled z-scores, and gene ontology enrichment analyses, supporting activation of a regeneration-associated transcriptional program in the mutants.
Sub-folder: 3B
Sub-sub folder: Raw (czi files)
Description: original czi files
Files: exported tif files
Description: named by animal number, genotype, age, and injury stage.
Sub-folder: 3C
Sub-sub folder: LENP2EGFP
Description: Represents genotype LENP2EGFP. Contains original czi files and exported tiff files, named by animal number and genotype.
Sub-sub folder: RENcfosEGFP
Description: Represents genotype RENcfosEGFP. Contains original czi files and exported tiff files, named by animal number and genotype.
Sub-folder: 3D
File: 3D EGFP positive area of RENcfosEGFP and LENP2EGFP.xlsx
File: 3D EGFP positive area of RENcfosEGFP and LENP2EGFP.csv
Description: quantified data from images in 3B and 3C in values of millimeters squared
File: 3D EGFP positive area at the tail tips of WT sib and fras1 KO with RENcfosEGFP or LENP2EGFP.prism
File: 3D EGFP positive area at the tail tips of WT sib and fras1 KO with RENcfosEGFP or LENP2EGFP.jpeg
Description: file with statistics and graphs
Sub-folder: 3E-H
Description: folders are empty as sequencing data and analysis were uploaded to GEO.
Folder: Figure 4
This figure demonstrates that fras1 mutant phenotypes are worsened when Fgf (fibroblast growth factor) signaling is inhibited using a heat shock–inducible dominant-negative Fgf receptor transgene (hsp70:dnfgfr1). Embryos were exposed to repeated heat shocks by incubation in 37°C egg water for 30 minutes (min) at 1, 2, and 3 dpf (days post-fertilization) and then analyzed at 4 dpf. Representative images compared hsp70:dnfgfr1 larvae, fras1 mutants, and their respective negative controls following the heat-shock regimen. Induction of the dominant-negative receptor blocked Fgf signaling and produced developmental defects that were mild in wild-type larvae but more severe in fras1 mutants. Measurements of body length showed that inhibition of Fgf signaling caused a modest reduction in larval size in both wild-type and fras1 mutant backgrounds. Analysis of morphology further revealed that fras1 mutants were more susceptible to developing a distorted posterior when Fgf signaling was disrupted. Together, these results indicate that Fgf signaling helps buffer developmental defects associated with fras1 loss and that reducing Fgf receptor activity exacerbates the mutant phenotype.
Sub-folder: Raw (czi files and exported tif files)
Description: original czi files exported tif files, with files named by animal number and genotype.
Other files: pptx and PDF file arranging the images of larvae with annotations. Spreadsheets are included as .csv files representing body length values in mm (Fig. 4D) and ratio of larvae showing phenotypes (Fig. 4E). A GraphPad statistical file with an exported graph image for 4D is also included.
Folder: Figure S1
Figure S1 summarizes regeneration-related phenotypes identified in F3 families from the ENU mutagenesis screen using EGFP (enhanced green fluorescent protein) reporter expression as a marker of regeneration-associated activity. Ten F3 families repeatedly exhibited regeneration phenotypes, including reduced sizes of regenerating tissues in families 1–5 and EGFP expression at fin fold edges in families 5–9, while yellow arrows indicate affected regions. The budlightmutants displayed the most severe phenotype, showing both strong regeneration defects and degeneration of fin folds even in the absence of injury. Images of larval phenotypes were collected with a 100 μm (micrometer) scale bar. Analysis of 25–35 screened F3 larvae per family showed that the proportion of affected larvae was generally consistent with recessive inheritance of the underlying mutations. Two of the identified families also produced adult fish with abnormal caudal fin regeneration, marked by defects visible at 2 dpa (days post-amputation), although regeneration appeared grossly normal by 7 dpa. White arrowheads indicate the fin amputation planes in adult analyses, and the adult regeneration images include a 1 mm (millimeter) scale bar.
Sub-folder: S1A
Description: contains 18 subfolders containing original czi files for each of embryos displaying phenotypes from the selected 18 F3 families, along with exported tif file, named by animal number. For each family, 3 different animals are captured.
File: S1B ENU phenotypic penetrance ratios.xlsx
File: S1B ENU phenotypic penetrance ratios.csv
Description: indicates ratios of larvae showing phenotypes from**** 25-35 F3 assessed larvae each from F2 mating (S1B).
Sub-folder: S1C
Description: contains four folders containing an original czi file and exported tif files with images of each adult regenerating caudal fin from F3 families 1 & 2, named by animal number and phenotype. There are 4 animals for F3 family 1, and 5 animals for F3 family 2.
Folder: Figure S2
Figure S2 quantifies fin fold degeneration and regeneration-associated reporter activity in budlight mutants using fgf20a:EGFP, where EGFP stands for enhanced green fluorescent protein. Fin fold area was measured from maximum-intensity projections of confocal z-stack images in larvae from 1–6 dpf (days post-fertilization). Wild-type larvae showed progressive fin fold growth throughout development, whereas budlight mutants reached a plateau in fin fold expansion by approximately 2 dpf, indicating impaired tissue growth. Fin fold size was reported in µm² (square micrometers), and each data point represented an individual larva with results presented as mean ± SEM (standard error of the mean). Reporter activity was quantified by measuring EGFP fluorescence intensity within the segmented fin fold and normalizing the signal to tissue area. This analysis revealed that budlight mutants exhibited elevated fgf20a:EGFP expression relative to wild-type controls beginning around 2–3 dpf, coinciding with the onset of fin fold degeneration. Together, these data show that impaired fin fold growth in budlight mutants is accompanied by increased activation of a regeneration-associated genetic program, as reflected by elevated area-normalized fgf20a reporter expression.
Sub-folder: S2A-B
Sub-sub-folder: Raw (czi files)
Description: contains the raw files for 6 WT and 6 BL (budlight) embryos across 1-6 dpf, which are named by genotype, age, and animal number.
Sub-sub-folder: Processed QC images
Description: contains TIFFs for CZI images with selected fin fold ROI area overlayed with brightfield/GFP signal, named by genotype, age, animal number, and channel
File: finfold_results.csv
Description: quantifications for fin fold area in square microns and GFP signal intensity in arbitrary units in the area.
File: S2A_plotDATA” and “S2B_plotDATA.xlsx
File: S2A_plotDATA” and “S2B_plotDATA..csv
Description: measurements (area in sq microns for S2A; arbitrary fluorescence units per measured area for S2B)
File: FigureS2_Statistics.xlx
File: FigureS2_Statistics.csv
Sub-folder: S2C
Description: contains original .czi and exported tiff images of wild-type and budlight mutant larvae assessed at 2 days post fertilization for tail blistering, named by genotype and animal number. Three animals of each genotype were analyzed.
Folder: Figure S3
Figure S3 summarizes a CRISPR-Cas9 F0 mutagenesis screen targeting human developmental disease-associated genes and their corresponding zebrafish orthologues to assess effects on regeneration-associated reporter activity. A table lists the human disease genes included in the screen along with the disorders associated with each gene. Control larvae expressing fgf20a:EGFP (enhanced green fluorescent protein) that were either uninjected, injected with Cas9protein alone, or injected with Cas9 plus a non-targeting guide RNA showed neither ectopic reporter activation nor obvious tissue degeneration. In contrast, CRISPR-Cas9 disruption of genes such as flnb and aspa induced ectopic fgf20a:EGFP expression within degenerating fin fold tissues, with yellow arrows marking affected regions. Fin fold morphology was quantified by measuring fin fold area and normalizing it to the mean control fin fold area for each experimental condition. Reporter activity was quantified as integrated EGFP fluorescence intensity normalized first to fin fold area and then to the average control value, with individual points representing single larvae and results presented as mean ± SEM (standard error of the mean). Statistical significance relative to controls was assessed using Welch’s t-tests with Benjamini–Hochberg FDR (false discovery rate) correction, where FDR < 0.05, 0.01, 0.001, and 0.0001 were denoted by *, **, ***, and ****, respectively.
Sub-folder: S3B
Description: contains processed images for the groups in part B, named by treatment and animal number. 20-30 larvae were imaged in a single tiling experiment for groups in S3B, with raw CZI files exceeding several GBs in size. Therefore, selected representative processed images are provided here, with raw files available upon request.
Sub-folders: S3C and S3D
Description: contain representative .jpg images selected for the figure, selected from those in the S3E-F folder, named by genotype and treatment.
Sub-folder: S3E-F
Description: contains the finfold_GFP_multiROI_results.csv file with quantifications for fin fold area in sq microns and GFP signal intensity in arbitrary units in the area. Also, present is the .ijm file which is the macro code that can be run with imageJ/FIJI to quantify images. A Raw CZI folder contains the raw files in .czi format for all gene specific injected and control larvae at 4 dpf, named by gene, animal age, and animal number. The Processed QC images folder contains TIFFs for CZI images with selected fin fold ROI area overlayed with GFP signal, named by gene, animal age, image number, and animal number.
File: finfold_GFP_multiROI_results.csv
Description: Primary measurement data. Per-embryo fluorescence and morphometric measurements from confocal images of fgf20a:EGFP reporter larvae at 4 dpf, following CRISPR targeting of 16 human congenital-disease orthologs plus uninjected control. Quantified as multiple regions of interest per image (1–5 embryos per field).
Variables:
- Condition (targeted gene or Control)
- Day (all day4)
- Image (source .czi filename)
- Selected_Channel (EGFP channel extracted for quantification)
- ROI_Number (embryo index within image)
- Animal_ID (unique per-embryo identifier)
- Area (fin fold area, px²)
- Mean_GFP (mean EGFP intensity, a.u.)
- Integrated_GFP (Area × Mean_GFP)
- GFP_per_Area (mean EGFP intensity per unit area, a.u.)
Note: This file is the source for all four derived files below.
File: S3E_plotDATA_Normalized_finFOLD_area.csv
Description: Plotting data for Supplementary Figure S3E. Fin fold areas from the file above, expressed as a ratio to the mean control area, in wide format with one column per condition. Columns are ragged: n differs by condition (7–21) and shorter columns are padded with empty cells.
File: S3E_plotDATA_STATISTICS.csv
Description: Statistical results underlying Supplementary Figure S3E. Sixteen pairwise Welch's t-tests comparing each targeted condition with control on normalized fin fold area, with Benjamini–Hochberg correction across the 16 comparisons. Columns: .y. and metric (response variable); group1/group2 (conditions compared); n1/n2 (embryos per group); statistic (t); df (Welch-adjusted, non-integer); p; p.adj; p.adj.signif and stars (significance codes); comparison (label).
File: S3F_plotDATA_Normalized_GFP_perAREA.csv
Description: Plotting data for Supplementary Figure S3F. EGFP intensity per unit area from the primary file, expressed as a ratio to the control mean, in the same wide, ragged format as S3E.
File: S3F_plotDATA_STATISTICS.csv
Description: Statistical results underlying Supplementary Figure S3F, for normalized EGFP per area. Same test, correction, and column structure as S3E_plotDATA_STATISTICS.csv.
Folder: Figure S4
Figure S4 outlines the RNA-seq (RNA sequencing) workflow used to compare gene expression in regenerating and mutant tissues, with tail amputations performed at 3 dpf (days post-fertilization) and samples collected from WT (wild-type) uninjured larvae at 4 dpf, WT regenerating tails at 1 and 2 dpa (days post-amputation), and age-matched fras1mutants. For each condition, tail tissues from 30–40 larvae were pooled for RNA isolation and non-directional library preparation. PCA (principal component analysis) of the transcriptomic data identified major sources of variation, and gene ontology analysis revealed that genes contributing to PC1 (principal component 1) were enriched for extracellular matrix (ECM) organization, connective tissue development, regeneration, and skeletal development pathways. In contrast, genes associated with PC2 (principal component 2) were enriched for developmental and sensory perception-related biological processes. The enrichment plots represented gene counts by dot size and statistical significance by color-coded adjusted p-values. Correlation analysis compared transcriptional changes in WT regeneration and fras1 mutants, including genes altered in one or both conditions while excluding genes unchanged in either condition. Genes were categorized as shared, WT regeneration-specific, or fras1-specific according to log2FC (log2 fold change) thresholds of |log2FC| ≤ 1 and FDR (false discovery rate)-corrected p-values < 0.05, with relationships quantified using Pearson and Spearman correlation coefficients and summarized by category frequencies.
Sub-folder: S4
Description: contains a .csv file “051826_Budlight_PROJECT_processedDATA_TPM” with Transcript per million values for all genes across the 8 samples. Also, it contains a bigwig folder containing individual bigwig files for IGV browser for each of the 8 samples.
Folder: Figure S5
Figure S5 demonstrates that skin and fin fold degeneration phenotypes in fras1 mutants are exacerbated by mild inhibition of Fgf (fibroblast growth factor) signaling using a heat shock–inducible dominant-negative Fgf receptor transgene (hsp70:dnfgfr1). Larvae were examined at 3 dpf (days post-fertilization) before heat shock and at 4 dpfafter heat shock treatment at 37°C for 30 min (minutes), showing that fin fold degeneration alone did not activate the hsp70 promoter in the absence of heat shock. A milder repeated heat-shock regimen was conducted at 2, 4, and 6 dpf, with phenotypic assessments performed at 2, 4, 6, and 8 dpf. Representative images revealed that fras1 mutant larvae carrying the hsp70:dnfgfr1 transgene displayed slightly smaller fin folds than fras1 mutants lacking the transgene, whereas wild-type siblings were largely unaffected. Quantitative measurements of body length and fin fold area showed no significant differences between wild-type siblings with or without hsp70:dnfgfr1, indicating that mild Fgf receptor blockade had minimal effects in a normal genetic background. In contrast, fras1 mutants exhibited measurable reductions in fin fold area when Fgf receptor signaling was inhibited, suggesting increased sensitivity to reduced Fgf activity. Data are presented as mean ± SEM (standard error of the mean) from n = 6 biological replicates, and statistical significance was assessed using two-sample t-tests assuming unequal variances or Fisher’s exact test with multiple-comparison correction, with P values reported for the relevant comparisons.
Sub-folder: S5A
Description: contain two subfolders with .czi files and tif files exported from czi files of brightfield and EGFP each for 6 larvae at 3 dpf before heat shock and 7 larvae at 4 dpf after heat shock. They all are fras1 mutants with hsp70:dnfgfr1 identified by skin degeneration and lens EGFP expression. Files are named by genotype, animal number, and imaging channel.
Sub-folder: S5C
Description: contains a jpeg file arranging the images with annotations, and a folder containing 16 subfolders with original czi files and exported tiff files each of WT siblings and fras1 KO with/without hsp70:dnfgfr1 at 2, 4, 6, and 8 dpf. There are 48 WT sibling animals and 48 fras1 mutant animals, and images are named by animal number.
Sub-folder: S5D
Description: contains a GraphPad prism statistical file “S5D body length of WT sib and fras1 KO with or without hspdnfgfr1” and exported jpeg graph image, and a spreadsheet “S5C - S5E WT sib and fras1 KO with or without hspdnfgfr1” in .xlsx and .csv formats with measurements of phenotyped larval length in millimeters.
Sub-folder: S5E
Description: contains a GraphPad prism statistical file “S5E finfold area of WT sib and fras1 KO with or without hspdnfgfr1“ and exported jpeg graph image, and a copy of the same spreadsheet as in Fig S5D “S5C - S5E WT sib and fras1 KO with or without hspdnfgfr1 copy“ in .xlsx and .csv formats with measurements of areas of phenotyped larval finfold area in sq millimeters.