The fluorescence in situ signal for zebrafish serpine3 is specific for wild type (WT, yellow), where it is in proximity to staining of the ZRF1-antibody (green). This indicates expression of serpine3 by Müller glia cells. Serpine3 signal is not present in homozygous serpine3^cbg17 siblings. In the overlay, nuclei (DAPI) are shown in white. ONL – outer nuclear layer, INL – inner nuclear layer, GCL – ganglion cell layer. Scale bar = 20 µm.

Fluorescence in situ hybridization of serpine3 mRNA (yellow) does not show co-localization with anti-protein kinase C alpha (PKCa encoded by prkca, green) or choline O-acetyltransferase a (CHAT, encoded by chata, green) proteins. In the overlay, nuclei (DAPI) are shown in white. ONL – outer nuclear layer, INL – inner nuclear layer, GCL – ganglion cell layer. Scale bar overview = 20 µm, details = 20 µm.

Sequenced reads are mapped against the danRer11 zebrafish genome assembly with Blat and are visualized in the UCSC genome browser (black) together with the serpine3 RefSeq annotation (blue). (A) PCRs with primers For1 and Rev2 or For1 and Rev2.2 (green) amplify the expected regions around the transcription start site on chromosome 9 in wild type (WT_Rev2, WT_Rev2.2). Injection of CRISPR guides 1, 2, and 3 (orange) results in a deletion of about 400 bp for founder individuals 7 and 12 in serpine3^cbg17. Offspring of founder 12 (394 bp deletion) was raised and further crossed. The location of a single serpine3 transcription start site is supported by annotation and activating histone marks H3K4me3 within the respective region (lower gray bar with darkness of color correlating with signal intensity). (B) PCR with primers For3 and Rev4 (green) amplifies a region around coding exon1 of serpine3. Injection of CRISPR guides 4, 5, and 6 (orange) results in a deletion of about 100 bp for founders 1, 9, 18, and 19 in serpine3^cbg18. For founder 9, this 92 bp deletion induces a frame-shift in the reading frame (bold) with three early stop codons (two shown as *, third stop located in coding exon 2) and offspring was grown. The deletion is equivalent to deletion of 31 amino acids (red) and +1 nt insertion. The amino acid encoded by a split codon is shown in blue.

(A) Genotyping of the serpine3^cbg17 with two primers (F0 generation). Injection of CRISPR guides into the one-cell embryo leads to mosaic embryos (several pooled at 72 hpf) with a strong wild type (WT) band at about 557 bp and a weak mutant band at about 161 bp, while uninjected embryos just have a single WT band (557 bp). The raw image is provided as Figure 4—source data 4. (B) Genotyping of serpine3^cbg17 with a mix of three primers (F2 generation). Heterozygous animals (H) have two bands, a WT (expected height: 286 bp) and mutant band (expected height: 161 bp), while homozygous mutants (M) and homozygous WT animals show a single band. NC – negative control (water). The raw image is provided as Figure 4—source data 5. (C) Genotyping of serpine3^cbg18 with two primers. Heterozygous animals (H) have two bands, a WT (expected height: 342 bp) and mutant band (expected height: 250 bp), while homozygous mutants (M) and homozygous WT animals show a single band. NC – negative control (water). The raw image is provided as Figure 4—source data 6.

Iris circularity is another descriptor for the deviation of eye shape and is defined as 4π × [Area]/[Perimeter]². A value of 1 indicates a perfect circle. Boxplots show that iris circularity significantly differs between WT (n=14) and serpine3^cbg17 eyes (n=10). The same holds for the comparison of WT (n=40) and serpine3^cbg18 eyes (n=40). A Wilcoxon Rank sum test was used. Boxplots display first quartile, median and third quartile with whiskers extending to the maximum and minimum within 1.5 times inter-quartile range. Outliers are shown in black.

No obvious difference in optic nerve morphology is observed between serpine3^cbg17 (A) and serpine3^cbg18 fish (B) and their respective wild type (WT) siblings. Scale bar = 200 µm.

We show details of representative overview images for one eye of each genotype on the right. In comparison to WT (A), distance between lens and retina of serpine3^cbg18 fish is reduced (distance bars). In the serpine3^cbg18 eye (B), all retinal layers are present and distinguishable, although they are not as tightly packed and clearly separated as in the WT (c–f). Moreover, we observed displaced pigmented cells that are located mainly in the photoreceptor outer segment and the outer nuclear layer yellow arrows, (c–e). Furthermore, the photoreceptor outer segment and RPE layer are not clearly separated in the serpine3^cbg18 retina. Scale bar in the overviews represents 200 µm and in the magnifications 20 µm.