Early redox, Src family kinase, and calcium signaling integrate wound responses and tissue regeneration in zebrafish

Abstract

Tissue injury can lead to scar formation or tissue regeneration. How regenerative animals sense initial tissue injury and transform wound signals into regenerative growth is an unresolved question. Previously, we found that the Src family kinase (SFK) Lyn functions as a redox sensor in leukocytes that detects H(2)O(2) at wounds in zebrafish larvae. In this paper, using zebrafish larval tail fins as a model, we find that wounding rapidly activated SFK and calcium signaling in epithelia. The immediate SFK and calcium signaling in epithelia was important for late epimorphic regeneration of amputated fins. Wound-induced activation of SFKs in epithelia was dependent on injury-generated H(2)O(2). A SFK member, Fynb, was responsible for fin regeneration. This work provides a new link between early wound responses and late regeneration and suggests that redox, SFK, and calcium signaling are immediate "wound signals" that integrate early wound responses and late epimorphic regeneration.

Figure 1.

Redox and SFK signaling at wounds. (A) Immunofluorescence of pSFK (phosphorylation of SFK activation loop tyrosine) in 3-dpf larvae at various time points. Arrows indicate the position of tail transection. The dotted line is the position of tail transection. (B) Immunofluorescence of pSFK and pErk in 3-dpf larvae at 30 min after tail transection. DPI and PP2 inhibit SFK activation at wounds but not Erk activation. (C) Immunofluorescence of pSFK and Cadherin in 2.5-dpf larvae at 30 min after tail transection. hpw, hour postwounding. Bars, 50 µm.

Figure 2.

Early redox and SFK signaling regulates late epimorphic regeneration. (A) Diagram of regeneration assays using zebrafish larval tail fins. (B) Quantification of regenerated tail fin length at 3 d after wounding (DMSO: 19 larvae; DPI: 20 larvae; PP2: 19 larvae). (C) Representative pictures used for quantification in B. Dotted lines indicate tail transection. (D) Quantification of blastemal proliferation at 36 h after wounding (DMSO: 19 larvae; DPI: 19 larvae; PP2: 21 larvae). Mitotic cells were detected using an antibody for phosphorylated histone H3 (H3P). (E) Representative pictures used for quantification in D. (F) Quantification of neutrophil recruitment to wounded fins at 3 d after wounding (DMSO: 30 larvae; DPI: 25 larvae; PP2: 25 larvae). (G) Representative pictures of Sudan black staining used for quantification in F. (H) Quantification of regenerated tail fin length at 3 d after wounding in control larvae and pu.1 MO-injected larvae (control/DMSO: 20 larvae; control/DPI: 13 larvae; control/PP2: 19 larvae; pu.1 MO/DMSO: 20 larvae; pu.1 MO/DPI: 20 larvae; pu.1 MO/PP2: 20 larvae). (I) Fluorescent pictures of 2.5-dpf Tg(mpx:Dendra2), which expresses Dendra2 specifically in neutrophils. (J) Sudan black staining of control larvae and pu.1 MO-injected larvae at 2.5 dpf. *, P < 0.05; one-way ANOVA with Dunnett’s posttest. Horizontal lines indicate means. Bars: (C, E, and G) 50 µm; (I and J) 100 µm.

Figure 3.

Wound-induced Ca²⁺ signaling is important for late regeneration. (A) Kymograph in the rectangular box in B. The rectangular box was summed into a 1D line, and the kymograph was made. (B) Time-lapse imaging of GCaMP3 in a 2-dpf larva immediately after wounding (see Video 1). (C) GCaMP3 images in DMSO- or thapsigargin-treated 2-dpf larvae (Video 2). (D) H₂O₂ imaging with HyPer in DMSO- or thapsigargin-treated 2-dpf larvae. Thapsigargin does not inhibit H₂O₂ burst at wounds. (E) Immunofluorescence of pSFK and pErk in DMSO- or thapsigargin-treated 2-dpf larvae. Thapsigargin did not inhibit pSFK or pErk at wounds. (F) Quantification of tail fin length at 3 d after treatment (+wound/DMSO: 23 larvae; +wound/thapsigargin: 32 larvae; +wound/U73122: 32 larvae; −wound/DMSO: 21 larvae; −wound/thapsigargin: 32 larvae; −wound/U73122: 30 larvae). Horizontal lines indicate means. (G) Representative pictures at 3 d after treatment. 2-dpf larvae were treated and wounded as described in Fig. 2 A. Dotted lines indicate tail transection. *, P < 0.05; one-way ANOVA with Dunnett’s posttest. Ex, excitation. Bars, 50 µm.

Figure 4.

Second injury induces regeneration in regeneration-defective wounds. (A) Diagram of the double-wounding assay. (B) Representative pictures of single-wounded larvae (a) and double-wounded larvae (b) at 3 d after wounding. Dotted lines indicate tail transection. (C) Quantification of regenerated tail fin length at 3 d after wounding (single wound/DMSO: 9 larvae; single wound/DPI: 14 larvae; single wound/PP2: 10 larvae; double wound/DMSO: 13 larvae; double wound/DPI: 15 larvae; double wound/PP2: 15 larvae). Horizontal lines indicate means. (D) Immunofluorescence of pSFK at 0.5 h after second wounding. *, P < 0.05; one-way ANOVA with Dunnett’s posttest. Bars, 50 µm.

Figure 5.

Fynb is responsible for epimorphic regeneration. (A) RT-PCR of SFKs. hck and lyn, hematopoietic-specific SFKs, are negative controls. (B) Tg(krt4:GFP) larvae at 3 dpf were used for flow cytometry. GFP high fractions (boxed with a blue line) were sorted. Cells from wild-type larvae were used to set the background level of autofluorescence. (C) RT-PCR of fynb and yes with/without MOs. ef1-α is the loading control. (D) Representative pictures at 3 d after wounding. (E) Quantification of regenerated tail fin length at 3 d after wounding (control [Ctrl]: 19 larvae; fynb MO1: 20 larvae; fynb MO2: 21 larvae; yes MO: 20 larvae; fynb MO1/yes MO: 16 larvae). (F) In situ hybridization of fynb mRNA in 2-dpf larvae. The tail fin in the blue box is magnified on the right. (G) Ratio images of pSFK/krt4-tdTomato in control and fynb morphants. (H) Mitotic cells were detected by antibody staining for phosphorylated histone H3 (H3P) at 36 h after wounding. (I) Quantification of blastemal proliferation at 36 h after wounding (control: 14 larvae; fynb MO1: 12 larvae). (J) A proposed model showing early wound signaling-mediated regulation of late regeneration. (E and I) *, P < 0.05; one-way ANOVA with Dunnett’s posttest (E) and two-tailed unpaired t test (I). Horizontal lines indicate means. Bars, 50 µm.