Analysis of zebrafish MZznf598 embryos and larvae.

(A) Brightfield images of wild-type and MZznf598 embryos at the 8-cell stage, shield stage, and 26 hpf. The scale bars indicate 200 μm. (B) Western blotting of phosphorylated eIF2α in wild-type and MZznf598 zebrafish embryos at 6 hpf. Total eIF2α and Tubulin (Tub) were detected as loading controls. (C) Brightfield images of wild-type and MZznf598 embryos at 9 dpf. Scale bars indicate 500 μm. (D) Distributions of body length in wild-type and MZznf598 fish at 2, 5, and 9 dpf. P values were calculated by the Mann–Whitney U test (two-tailed). The numerical data underlying this figure can be found in S1 Data. dpf, days postfertilization; hpf, hours postfertilization.

Transcriptome analysis of MZznf598 embryos during the MZT.

(A) A scheme of RNA-Seq analysis comparing wild-type and MZznf598 embryos during the MZT. (B) Venn diagrams of mRNAs up-regulated or down-regulated at 1 hpf or 6 hpf (FDR < 0.01). The numbers of genes in each category are shown. (C) Functional enrichment analysis of biological terms associated with genes up-regulated (orange) or down-regulated (light blue) at 6 hpf. Terms that were significantly enriched (FDR < 0.01) are shown. (D) A volcano plot showing the fold change in mRNA expression between wild-type and MZznf598 embryos at 6 hpf (x-axis) and FDR (y-axis). Genes encoding C2H2-ZF (red) and other genes (gray) are shown. (E) Bar graph showing the fractions of C2H2-ZF or RING-ZF genes among all the genes (gray), genes up-regulated in MZznf598 (orange), and genes down-regulated in MZznf598 (blue) at 6 hpf. The p value was calculated by the chi-square test. (F) Cumulative distributions of fold changes in mRNA levels between MZznf598 embryos and wild-type embryos at 6 hpf. All genes (black), C2H2-ZF genes (red), and RING-ZF genes (purple) are shown. The x-axis shows the fold change, and the y-axis shows the cumulative fraction. The p values are shown on the right (Kolmogorov–Smirnov test). (G) Bar graphs summarizing the expression patterns of C2H2-ZF genes (red) and other genes (gray) up-regulated in MZznf598 embryos at 6 hpf. The numerical data underlying this figure can be found in S1 Data. hpf, hours postfertilization; MZT, maternal-to-zygotic transition.

Validation of maternal mRNAs up-regulated in MZznf598.

(A) qRT-PCR analysis of maternal NGD target candidate mRNAs in wild-type (blue) and MZznf598 embryos (red) at 6 hpf relative to 2 hpf. The results for MZznf598 embryos rescued by injecting mRNAs encoding Myc-tagged full-length Znf598 (light blue) or a mutant Znf598 lacking the RING domain (pink) are also shown. gstm mRNA (miR-430 target) and ddx4 mRNA (stable maternal mRNA) are shown as controls for Znf598-independent maternal mRNAs. (B) Western blotting to detect Myc-tagged Znf598 proteins at 6 hpf. Tubulin (Tub) was detected as a loading control. (C) Time course qRT-PCR analysis of znf236 and znf970 mRNAs in wild-type (blue) and MZznf598 (red) embryos. (D) Schematics of the Znf236 and Znf970 proteins. C2H2-ZF is indicated in orange. (E) qRT-PCR analysis of znf236 mRNA in wild-type and MZznf598 embryos at 6 hpf injected with control MO (blue and red) or translation-blocking MO (green and orange). (F) PAT assay of znf236 mRNA in wild-type and MZznf598 embryos at 6 hpf. The lane labeled A0 shows the 3′ UTR fragment without a poly(A) tail. (G) Quantification of the PAT assay in (F). The average values of the 2 experiments are shown. The graphs represent the average of 3 independent experiments in A, C, and E. The error bars show the standard deviation. The open circles in A and E show each data point. The numerical data underlying this figure can be found in S1 Data. hpf, hours postfertilization; MO, morpholino oligonucleotide; NGD, no-go decay.

Validation of zygotic C2H2-ZF mRNAs up-regulated in MZznf598.

(A) Cumulative distributions of fold changes in mRNA levels in MZznf598 embryos compared to wild-type embryos at 6 hpf. C2H2-ZF genes on chromosome 4 (orange), C2H2-ZF genes on other chromosomes (purple), and genes without C2H2-ZF (black) are shown. The x-axis shows the fold change, and the y-axis shows the cumulative fraction. The p values are shown on the left (Kolmogorov–Smirnov test). (B) qRT-PCR analysis of C2H2-ZF mRNAs in wild-type (blue) and MZznf598 embryos (red) at 6 hpf. The results for MZznf598 embryos rescued by injecting mRNAs encoding Myc-tagged full-length Znf598 (light blue) or a mutant Znf598 lacking the RING domain (pink) are also shown. actb1 mRNA is shown as a control for Znf598-independent mRNA. The chromosomal locations are indicated below the gene name. (C) A scheme of typical FZNF mRNA and corresponding pre-mRNA/genetic loci. (D) Genomic qPCR analysis using primers that amplify exons encoding the FiNZ domain. The estimated copy number of FiNZ exons per haploid genome normalized to that of rpl24 is shown on the y-axis. (E) qRT-PCR analysis of FZNF mRNAs and pre-mRNAs at 6 hpf. The colors are the same as in (B). (F) qRT-PCR analysis of FZNF mRNAs at 6 hpf in wild-type embryos injected with control MO (blue), znfMO4 (light green), or FiNZ ATGMO (green). znf292b (a C2H2-ZF gene without the FiNZ domain) and actb1 (a gene without C2H2-ZF) are shown as controls. The graphs in B, E, and F represent the average of 3 independent experiments. The graphs in D represent the average of 8 independent embryos. The error bars show the standard deviation. The open circles in B, D, E, and F show each data point. Asterisks in E and F indicate p < 0.05 (Dunnett’s test, compared to wild type or control MO). The numerical data underlying this figure can be found in S1 Data. hpf, hours postfertilization; MO, morpholino oligonucleotide.

Analysis of ribosome stalling sequences in C2H2-ZF mRNAs.

(A) A schematic of the tandem ORF assay. The Rluc ORF (orange) and Fluc ORF (light green) are separated by 2 P2A translation skipping sequences (blue). The sequence to be tested (shown as X in gray) is inserted between the 2 P2A sequences. (B) Schematic of the znf236 ORF fragments analyzed in this study. C2H2-ZF is indicated in orange. Fragments that caused significant ribosome stalling in the tandem luciferase assay are indicated in red. (C, D) Results of the tandem ORF assay with the znf236 fragment. (E) Western blotting to detect Myc-tagged EGFP (left panel) or Myc-tagged Znf236 fragment 1 in the wild type and MZznf598 at 6 hpf. Tubulin (Tub) was detected as a loading control. The predicted positions of Myc-tagged peptides truncated in the Znf236 1A-1D regions are indicated on the right. (F) A detail of the 1B fragment. The amino acid sequences of ZF6, ZF7, and ZF8 are shown below. A-site positions of the stalled leading ribosome and corresponding footprint amounts are indicated as a red gradient. (G) Results of the tandem ORF assay with ZF6, ZF7, or ZF8 repeats. (H) Results of the tandem ORF assay with different numbers of ZF6 repeats. (I) Distributions of disome footprints (blue) in the znf1102 and si:ch73-266f23.1 ORFs. The amino acid sequences of C2H2-ZFs with high disome footprint amounts are shown below. A-site positions of the stalled leading ribosome and corresponding footprint amounts are indicated as a red gradient. (J) Results of the tandem ORF assay with 1B fragment (ZF5-6-7-8) of znf236, ZF4-5-6-7 of znf1102, and ZF3-4-5 of si:ch73-266f23.1 in wild-type (blue) and MZznf598 (red) embryos. The graphs in C, D, G, H, and J represent the average of 3 independent experiments. Relative Fluc signals normalized to Rluc signals are shown. Values with no insert (-) were set to one. The error bars show the standard deviation. The open circles show each data point. Black asterisks indicate p < 0.05 compared to no insert. Red asterisks indicate p < 0.05 compared to LacZ-N1650 or GFP (Dunnett’s test). Daggers indicate p < 0.05 (Student’s t test). The numerical data underlying this figure can be found in S1 Data. hpf, hours postfertilization.

Analysis of the effect of C2H2-ZF sequences on mRNA stability.

(A) A scheme of the mRNA injection assay used to measure the effect of the sequence to be tested (shown as X in gray) on mRNA stability. (B) qRT-PCR analysis of injected sfGFP reporter mRNAs at 6 hpf relative to 2 hpf in wild-type (blue) and MZznf598 (red) embryos. (C) qRT-PCR analysis of injected sfGFP-ZF6×3 reporter mRNAs at 6 hpf relative to 2 hpf in wild-type (blue), MZznf598 (red), GFP MO-injected wild-type (light green), and GFP MO-injected MZznf598 (orange) embryos. The graphs in B and C represent the average of 3 independent experiments. The error bars show the standard deviation. The open circles show each data point. Asterisks indicate p < 0.05 (Student’s t test in B and Dunnett’s test compared to the wild type in C). (D) A scheme of the disome localization score analysis. (E) Cumulative distributions of fold changes in mRNA levels in MZznf598 embryos compared to wild-type embryos at 6 hpf. All genes (black) and C2H2-ZF genes with high (dark blue), middle (blue), and low (turquoise) disome localization scores are shown. The x-axis shows the fold change, and the y-axis shows the cumulative fraction. The p values are shown on the right (Kolmogorov–Smirnov test). (F) A model of NGD for degrading mRNAs encoding stall-prone tandem C2H2-ZF sequences in zebrafish. The numerical data underlying this figure can be found in S1 Data. hpf, hours postfertilization; MO, morpholino oligonucleotide; NGD, no-go decay.