The ezh2(sa1199) mutant zebrafish display no distinct phenotype

Abstract

Polycomb group (PcG) proteins are essential regulators of epigenetic gene silencing and development. The PcG protein enhancer of zeste homolog 2 (Ezh2) is a key component of the Polycomb Repressive Complex 2 and is responsible for placing the histone H3 lysine 27 trimethylation (H3K27me3) repressive mark on the genome through its methyltransferase domain. Ezh2 is highly conserved in vertebrates. We studied the role of ezh2 during development of zebrafish with the use of a mutant allele (ezh2(sa1199), R18STOP), which has a stop mutation in the second exon of the ezh2 gene. Two versions of the same line were used during this study. The first and original version of zygotic ezh2(sa1199) mutants unexpectedly retained ezh2 expression in brain, gut, branchial arches, and eyes at 3 days post-fertilization (dpf), as revealed by in-situ hybridization. Moreover, the expression pattern in homozygous mutants was identical to that of wild types, indicating that mutant ezh2 mRNA is not subject to nonsense mediated decay (NMD) as predicted. Both wild type and ezh2 mutant embryos presented edemas at 2 and 3 dpf. The line was renewed by selective breeding to counter select the non-specific phenotypes and survival was assessed. In contrast to earlier studies on ezh2 mutant zebrafish, ezh2(sa1199) mutants survived until adulthood. Interestingly, the ezh2 mRNA and Ezh2 protein were present during adulthood (70 dpf) in both wild type and ezh2(sa1199) mutant zebrafish. We conclude that the ezh2(sa1199) allele does not exhibit an ezh2 loss-of-function phenotype.

Fig 1. Validation of ezh2(sa1199) genotypes.

A. Ezh2, its domains, and mutant allele positions (grey lines). The green and red boxes indicate WD-binding and SET domains, respectively. The ezh2(sa1199) allele (left) has a stop mutation on the arginine 18 position (R18STOP), ezh2(ul2) allele (middle) has a 22 bp insertion that leads to a nonsense codon at amino acid 60, and ezh2(hu5670) allele (right) has a nonsense mutation on the arginine 592 (R592STOP) position. B. Allele specific (top) and PCR-restriction (bottom) genotyping on caudal fin clips of 2 dpf embryos. Agarose gel electrophoresis shows differential amplification or restriction enzyme digestion of the alleles. A wild type, heterozygous, and mutant genotype from a single experiment is represented for each method. C. Allele-specific genotyping validation by Sanger sequencing. Two dpf embryo fin clips were allele-specifically genotyped and Sanger-sequenced to validate the genotyping method. The mutation locus (black box) is visualized for wild type (left), heterozygous (middle), and mutant (right) embryos. D. Allele specific genotyping of 1, 2, and 3 dpf ezh2(sa1199) in-crossed embryos (N = 17 per day). The percentage of wild types (black), heterozygotes (grey), and mutants (brown) show a Mendelian ratio. The number of embryos is indicated inside the bars of the graph in white.

Fig 2. ezh2(sa1199) zebrafish line has non-specific phenotypes.

A. Whole mount in-situ hybridization for ezh2 on 3 dpf ezh2(sa1199) (left panel) and ezh2(hu5670) (right panel) embryos. Left panel. Wild type siblings (top) show ezh2 expression in the brain, eyes, branchial arches, and the gut (white arrowheads). Heterozygotes (middle) and ezh2(sa1199) mutants (bottom) show the same expression pattern; they are phenotypically comparable to wild types and have the same spatiotemporal ezh2 expression pattern. Right panel. Wild type siblings (top) show similar ezh2 expression to ezh2(sa1199) in-cross embryos (white arrowheads). Heterozygotes (middle) show decreased expression, and ezh2 is barely detectable in ezh2(hu5670) mutants (bottom). Scale bar: 500 μm. Numbers indicate the number of embryos displaying the shown expression pattern per total number of embryos tested. B. Left panel shows normal wild type and mutant embryos at 3 and 2 dpf, respectively. The right panel shows 2 and 3 dpf wild type and mutant embryos, respectively, with yolk sac, heart, and brain edemas (white arrowheads) and spinal curvatures (black arrowheads).

Fig 3. Selectively out-crossed ezh2(sa1199) survives until adulthood.

A. Survival graph of ezh2(sa1199). Survival assay was performed with ezh2(sa1199) in-crossed wild type (N = 18, blue line) and mutant (N = 21, orange line) siblings until 46 dpf. Some larval lethality during raising was seen both in wild types and mutants. The graph indicates that the survival of homozygous ezh2(sa1199) mutants is not significantly different from wild type siblings. B. Pictures of wild type and mutant ezh2(sa1199) adult siblings. While wild types have normal jaws (left), adult ezh2(sa1199) mutants (right) display an open mouth phenotype (black arrowhead). C. RT-qPCR measurement of ezh1 and ezh2 mRNA in ezh2(sa1199) adults. The (decapitated) body of 70 dpf wild type (blue) and mutant (orange) individuals were lysed and the presence of ezh1 and ezh2 mRNA was quantified and normalized to the reference genes ß-actin and ef1a. The ezh1 (p = 0.9509) and ezh2 (p = 0.1493) mRNA levels were not significantly altered in mutants. D. Ezh2 protein in single 70 dpf adults and H3K27me3 in 7 dpf pooled larvae. The presence of Ezh2 was tested in wild type (N = 3, left) and mutant (N = 4, right) bodies at 70 dpf, where Ezh2 expression persists (upper panel). At 7 dpf, mutant larvae (right) show 41.8% decreased levels of the H3K27me3 mark compared to wild type (left) larvae (lower panel) and relative to β-actin loading control. For each Western blot, representative bands were cropped from the same blot and shown in boxes for clarity of presentation.