Interferon signaling promotes early neutrophil recruitment after zebrafish heart injury

Figure 1:. Interferon signaling is induced by zebrafish heart injury. A) A time course of ifnphi1 gene expression after cryoinjury (2 to 72 hpi) normalized to elf1a expression. Relative fold change calculated by normalizing to uninjured control ventricles. B) ifnphi1–4 gene expression in wildtype ventricles at 2 hpi relative to uninjured ventricles. C) A time course of ifnphi1 gene expression after sham injury (2 to 72 hpi) relative to uninjured ventricles. D) A time course of isg15 gene expression after cryoinjury (2 to 120 hpi) relative to uninjured ventricles. Each dot shown represents a single ventricle after dissection away from the atrium and bulbus arteriosus. Data are shown as the mean ± s.d.

Figure 2:. A zebrafish mutant in ifnphi1 disrupts interferon signaling after heart injury. A) A design for mutating the ifnphi1 gene, with flanking guide RNAs shown (above) and the ifnphizj7 allele with a large deletion that removes part of exon 1, all of exon 2, and part of exon 3 (see Table S1 for sequences). B) A time course of isg15 gene expression after cryoinjury in ifnphi1−/− ventricles (2 to 120 hpi) relative to uninjured ventricles. C) Representative images of RNA fluorescence in situ hybridization (RNA-FISH) of isg15 (interferon stimulated gene-15) and myl7 (labels cardiomyocytes) in wild-type and ifnphi1−/− ventricle cryosections at 3 dpi. Sections were counterstained with DAPI (blue). D) Percent isg15 calculated via the number of isg15 speckles quantified in CellProfiler divided by the number of nucleated cells in the injured and uninjured zones. Welch’s t-test was used to determine statistical significance between the isg15 speckle percent in the injured zones (p = 0.0023). The dots on the graph represent the uninjured or injured zones of individual ventricles. Data are shown as the mean ± s.d.; n=4 wild-type and n=6 ifnphi1−/− ventricles. Scale bar = 50 μm. E) ifnphi1–4 gene…

Figure 3:. Interferon signaling promotes neutrophil recruitment to injured myocardium. A) Representative images of wild-type Tg(mpx:GFP) and ifnphi1−/− Tg(mpx:GFP) heart sections at 1 dpi. Sections were counterstained with DAPI (blue). Neutrophils are shown in white. B) Quantification of the percent of neutrophils and absolute number of neutrophils in the injured and uninjured zones of wild-type and ifnphi1−/−hearts at 1 dpi. The total number of neutrophils was counted in the uninjured and injured zones in wild-type and ifnphi1−/− hearts (p=0.0137). The percent of neutrophils was calculated by dividing the number of neutrophils by the number of nucleated cells in the uninjured and injured zones in wild-type and ifnphi1−/− hearts (p=0.0012). C) Representative images of wild-type Tg(mpx:GFP) and ifnphi1−/− Tg(mpx:GFP) heart sections at 3 dpi. Sections were counterstained with DAPI (blue). D) Quantification of the percent of neutrophils (p=0.0590) and absolute number of neutrophils (p=0.0157) in the injured and uninjured zones of wild-type and ifnphi1−/− hearts at 1 dpi was calculated as above. Each dot on the graphs represents the uninjured or injured zones calculated from individual sections…

Figure 4:. Interferon signaling is not required for macrophage recruitment to the heart injury zone. A) Representative images of RNA-FISH of mpeg (labels macrophages) and myl7 (labels cardiomyocytes), in wild-type and ifnphi1−/− ventricle cryosections at 3 dpi. Sections were counterstained with DAPI (blue). B) Percent macrophages were calculated by dividing the number of mpeg speckles by the number of nucleated cells in the injured and uninjured zones. The dots on the graph represent the uninjured or injured zones calculated from individual sections of distinct heart ventricles. Data are shown as the mean ± s.d.; n=7 wild-type and n=9 ifnphi1−/−. Scale bar = 50 μm.

Figure 5:. Interferon signaling is not required for neovascularization after heart injury. A, C, F) Representative images of RNA-FISH of kdrl (labels vasculature) and myl7 (labels cardiomyocytes) in uninjured, 3 dpi, and 7 dpi wild-type and ifnphi1−/− ventricle cryosections. Sections were counterstained with DAPI (blue). B, D, G) Percent kdrl was calculated by dividing the number of kdrl speckles by the number of nucleated cells in the injured and uninjured zones at each time point. E, H) kdrl enrichment was calculated by dividing the normalized kdrl speckle count in the injured zone by the normalized kdrl speckle count in the uninjured zone of the same heart section, for each heart. The dots on the graph represent the uninjured or injured zones calculated from individual sections of distinct heart ventricles. Data are shown as the mean ± s.d.; n=5 wild-type and n=5 ifnphi1−/− for uninjured samples, n=4 wild-type and n=7 ifnphi1−/− for 3 dpi samples, n=6 wild-type and n=4 ifnphi1−/− for 7 dpi samples. Scale bar = 50 μm.

Figure 6:. ifnphi1 mutant hearts are modestly impaired at resolving scars after injury. A) Experimental design for heart injury and collection at 30 and 60 dpi. B) Images of AFOG-stained sections of wild-type and ifnphi1−/−hearts at 30 and 60 dpi, showing representatives of mild and severe categories of scar resolution. C,D) Quantification of scar size (See Methods) in AFOG-stained hearts between wild-type and ifnphi1−/− stained sections at each time point. (E,F) Scar severities for all 30 dpi and 60 dpi injured hearts were categorized into three groups: fully regenerated (0–2% collagen), mildly scarred (2–5% collagen), or severely scarred (greater than 5% collagen). Chi-squared test for categorical data showed that 30 dpi categories between wild-type and ifnphi1−/− are not significantly different (p=0.1517), while 60 dpi categories between wild-type and ifnphi1−/− are modestly statistically significant (p=0.0378). The dots on the graph represent the uninjured or injured zones calculated from individual sections of distinct heart ventricles. Data are shown as the mean ± s.d.; n=19 wildtype and n=20 ifnphi1−/− for 30 dpi samples, n=15 wildtype and n=20 ifnphi1−/− for 60 dpi samples. Scale…