(A-D) Stereo and confocal microscope images of regenerating fins in wildtype (A,C) and mdka^mi5001 (B,D), immunolabeled with PCNA (magenta). Red lines (A,B) indicate the plane of cross section shown in panels C and D. White lines across the fins show the levels of the amputation plane (C,D). Scale bars equal 400 μm in (A,B) and 50 μm in (C,D). (E-H) Confocal images of the blastemal compartments of fins from wildtype (E,G) and mdka^mi5001 (F,H), immunolabeled with PCNA (E,F) or BrdU (G,H) incorporated between 3–4 dpa (see Materials and Methods). Scale bars equal 50 μm in (E-H). (I) Graph illustrating the average number of BrdU-positive cells within a 400 mm linear distance. wildtype: 186.8 ± 15.1 cells, mdka^mi5001: 142.9 ± 12.3 cells, Student’s t-test, p = 0.0326, n = 5 wildtype and 5 mutants.

(A) Fins from wildtype (top) and mdka^mi5001 (bottom) at 0, 4, 6, 19, and 32 dpa. Red lines indicate the level of the amputation plane. In wildtype animals, the structure of the bony fin rays and pigmentation recover completely. Compared to age-matched group of wildtype, pre-amputated fin in the mdka^mi5001 is smaller in size (wildtype: 51.3 ± 7.1 μm²; mdka^mi5001: 34.3 ± 4.7 μm², p<0.0001, Student’s t-test). In the mutants, regenerated fins display abnormal shape and mis-patterned melanophore pigmentation (red arrow). Scale bar equals 400 μm. (B) Graph illustrating the average percent outgrowth through 32 dpa, calculated from the ratio of the areas of pre- and post-amputation fins. Wildtype fins recover to 100% of pre-amputation levels by 32 dpa. The mutant animals show a significant reduction of fin outgrowth at 4 dpa. At the subsequent time points, there is no statistically significant difference in the fin outgrowth between wildtype and mdka^mi5001. (C) Graph illustrating the percent fin outgrowth at 4 dpa in wildtype and mdka^mi5001. Student’s t-test, p = 0.0105. n = 9 wildtype and 8 mutants.

(A) qPCR for midkine-a in wildtype zebrafish at 48 hours post lesion. Student’s t-test, p*<0.05, n = 3 uninjured and 3 injured (10 muscles per sample). (B) Cell counts of EdU labeled cells in muscles at 24 hours post lesion from wildtype (n = 4) and mdka^mi5001(n = 4). Student’s t-test, p*<0.05. (C) Diagram of the craniectomized head of an adult zebrafish. Red lines identify the lateral rectus muscles. Excised muscle is represented as red dashed line. Asterisk identifies the pituitary body at the midline. Boxes represent areas illustrated in panels D and E. (D) EdU-labeled cells in muscles from wildtype (top) and mdka^mi5001 (bottom) at 24 hours post lesion. High magnification inset image illustrates an elongated myonucleus undergoing proliferation (also see [13]). Arrows indicate the growing tip of the regenerating muscle. Asterisk indicates the pituitary body at the midline. (E) Extraocular muscles are outlined by dashed lines in control animals and in mutants at 4 days post lesion. Asterisks identify the pituitary body. (F) Outgrowth of regenerating muscles at post-injury time points in wildtype and mdka^mi5001. One-way ANOVA followed by Newman-Keuls multiple comparisons test, p*<0.05, p**<0.01, n = 4 wildtype and 4 mutants.

(A-D) Immunocytochemistry for proliferative marker, PCNA, in wildtype (A,B) and mdka^mi5001 (C,D) at post-lesion time points. (E-G) Graphs illustrating the number of PCNA-positive cells in the outer nuclear layer (E; 48 hpl, wildtype: 20.9 ± 3.7 cells, mdka^mi5001: 2.6 ± 2.1 cells, p = 0.0016; 72 hpl, wildtype: 18.5 ± 5.8 cells, mdka^mi5001: 3.8 ± 1.1 cells, p = 0.0126), inner nuclear layer (F; 48 hpl, wildtype: 50.7 ± 11.9 cells, mdka^mi5001: 4.4 ± 5.3 cells, p = 0.0036; 72 hpl, wildtype: 43.7 ± 9.7 cells, mdka^mi5001: 8.9 ± 7.3 cells, p = 0.0077), and total retinal layer (G; 48 hpl, wildtype: 72.1 ± 12.9 cells, mdka^mi5001: 7.1 ± 6.3 cells, p = 0.0014; 72 hpl, wildtype: 56.4 ± 19.1 cells, mdka^mi5001: 12.8 ± 8.0 cells, p = 0.0212). Scale bar equals 60 μm. Student’s t-test, p*<0.05, n = 3 wildtype and 3 mutants. ONL: outer nuclear layer; INL: inner nuclear layer.

(A, B) Immunocytochemistry for the cell marker, HuC/D, in wildtype (A) and mdka^mi5001 (B) at 72 hpi. (C) The number of HuC/D-positive cells following the injection of PBS (wildtype: 241.3 ± 28.2 cells, mdka^mi50011: 206.0 ± 12.4 cells) and ouabain (wildtype: 25.8 ± 10.5 cells, mdka^mi5001: 17.2 ± 7.2 cells, p = 0.137) retinas. (D, E) Immunocytochemistry for proliferative marker, PCNA in wildtype (D) and mdka^mi5001 (E) at 72 hpi. (F) The number of PCNA-positive cells per 500 μm in wildtype (54.4 ± 14.2 cells) and mdka^mi5001 (17.1 ± 6.4 cells. p = 0.00246) at 72 hpi. Student’s t-test, p*<0.05, n = 9 wildtype and 9 mutants. ONL: outer nuclear layer; INL: inner nuclear layer.

(A, B) BrdU-labeled regenerated retinal neurons in wildtype (A) and mdka^mi5001 (B) retinas at 28 dpi. Arrows indicated BrdU-labeled rod photoreceptors. (C) The number of BrdU-positive cells wildtype (125.2 ± 37.1 cells) and mdka^mi5001 (33.2 ± 14.4 cells, p = 0.000341) retinas at 28 dpi. (D,E) Regenerated inner neurons labeled with HuC/D in wildtype (D) and mdka^mi5001 (E) at 28 dpi. Arrowheads indicate regenerated inner neurons are displaced in the inner plexiform layer. (F) The number of HuC/D cells in wildtype (224.8 ± 74.6 cells) and mdka^mi5001 (76.8 ± 61.4 cells, p = 0.00021) at 28 dpi. Student’s t-test, p*<0.05, n = 9 wildtype and 9 mutants. ONL: outer nuclear layer; INL: inner nuclear layer.

(A,B) Following fin amputation, Midkine-a regulates proliferation of precursors produced from dedifferentiation of osteoblasts and activation of mesenchymal stem and progenitors (A). Following extraocular muscle excision, Midkine-a regulates proliferation of precursor produced from dedifferentiation of myocyte (B). During fin and muscle regeneration, loss of Midkine-a results in diminished initial burst of proliferation. (C) During regeneration of retinal neurons, Midkine-a governs proliferation of dedifferentiated Müller glia. In the absence of Midkine-a, retinal neurons fail to regenerate.