Fig 2: miR-216a and Dot1l Regulate Retinal Regeneration through the Wnt/β-Catenin Pathway(A) Control MOs or Dot1l MOs were injected and electroporated into the left eyes of wild-type zebrafish, followed by either 4% DMSO or GSK-β-inhibitor (1 mM) before intense light exposure (0 h). Eyes were collected after 51 h of intense light lesioning and immunostained using antibodies against PCNA. Nuclei were counterstained with TOPRO (blue).(B) dot1l MOs significantly decreased the number of PCNA+ cells in the INL, while there was no significant difference between dotl1l MO + GSK-3β-inhibitor co-injected eyes and control eyes.(C) Quantification of PCNA+ proliferating progenitors in MO-ctl + DMSO, MO-dot1l + DMSO, MO-ctl + GSK-3β-inhibitor, and MO-dot1l + GSK-3β-inhibitor electroporated retinas. Activation of Wnt signaling rescued the decrease in the number of proliferating progenitors upon dot1l knockdown after 51 h of intense light lesioning. Data represent means ± SEMs; n = 9–11 fish. *p < 0.05, p = 0.0167 (MO-ctl + DMSO versus MO-dot1l + DMSO) by 1-way ANOVA with Dunnett’s multiple comparisons test.(D) GSK-3β-inhibitor alone, Dot1l inhibitor (iDot1l) alone, or the combination was injected into the left eyes of Tg(1016tuba1a:GFP) zebrafish; DMSO alone was used as a control. At 51 h post-injection, eyes were collected for PCNA immunostaining. GSK-3β-inhibitor alone induced MG proliferation in undamaged eyes while co-injection of the Dot1l inhibitor led to no significant changes in number of proliferating MG. Data represent means ± SEMs. Each data point represents an individual fish. ****p < 0.0001, by 1-way ANOVA with Dunnett’s multiple comparisons test.(E) Control MO or miR-216a MO was injected and electroporated into the left eyes of Tg(1016tuba1a:GFP) zebrafish, followed by either 4% DMSO or XAV939 (10 µM). Eyes were collected at 51 h post-injection and immunostained using antibodies against GFP for dedifferentiated MG and PCNA for proliferating progenitors. Nuclei were counterstained with TOPRO (blue).(F) Suppression of miR-216a by MO-216a injection stimulates MG proliferation; however, upon co-injection with XAV939, no significant increase in the number of proliferating progenitors was detected.(G) Quantification of PCNA+ proliferating progenitors in MO-ctl + DMSO, MO-216a + DMSO, MO-ctl + XAV939, and MO-216a + XAV939 electroporated retinas. Inhibition of Wnt signaling reversed the increase in the number of progenitors upon miR-216a knockdown after 51 h of intense light lesioning. Data represent means ± SEMs; n = 18–21 fish. **p < 0.01, p = 0.0089 (MO-ctl + DMSO versus MO-216 + DMSO) by 1-way ANOVA with Dunnett’s multiple comparisons test.GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer. Scale bars, 50 µm.See also Figure S4.

Fig 3: miR-216 Targets Dot1l in the Retina during Photoreceptor Regeneration(A) For post-mitotic MG isolation, GFP+ cells were sorted from dark adapted undamaged Tg(gfap:gfp) retinas. For dedifferentiated MG isolation, GFP+ cells were isolated from 45-h light-lesioned Tg(1016tuba1a:GFP) retinas.(B) Fold changes in dot1l levels in FACS-purified MG were determined by qPCR. After 45 h of light damage, dot1l is upregulated in dedifferentiated MG (GFP+) in Tg(1016tuba1a:gfp) fish. Dot1l expression did not change in non-MG cells (GFP-) during regeneration. Data represent the means ± SEMs from 15 undamaged fish, and dedifferentiated MG were purified from 18 light-damaged fish.(C) Experimental scheme to test the effects of miR-216a overexpression on dot1l levels. Wild-type adult zebrafish were dark adapted and then either control miRNA or miR-216a was injected and electroporated into the left eyes before intense light exposure. After 24 h of light exposure, retinas were dissected for RNA isolation.(D) Fold changes in miR-216a levels in control miRNA (small interfering RNA [siRNA] against luciferase [siLuc]) or miR-216a mimic electroporated retinas were quantified by qPCR. miR-216a levels were upregulated by 15-fold in miR-216a mimic-injected retinas compared to controls. Data represent the means ± SEMs from 6 retinas.(E) Fold changes in dot1l levels in siLuc or miR-216 mimic electroporated retinas were quantified by qPCR. After 24 h of light damage, dot1l was downregulated in miR-216a-overexpressing retinas ~20-fold. Data represent means ± SEMs from 3 independent experiments. Six retinas were pooled for RNA isolation in each experiment.**p < 0.01 (Student’s t test), p = 0.0093.Scale bars, 50 µm.See also Figures S1 and S2.

Fig 4: Dot1l Is a Direct Target of miR-216a(A) Schematic of the reporter mRNA consisting of the coding sequence of GFP fused to the dot1l 3′ UTR. Three predicted miRNA recognition elements (MREs) are indicated. Predicted base pairing between MREs (shown in green) and the miR-216a sequence (shown in red).(B) Embryos injected at the 1-cell stage with 100 pg of GFP-dot1l 3′ UTR, with or without 100 pg miR-216a, were examined for GFP expression at 1 day post fertilization (dpf). GFP expression was apparent in embryos injected with GFP-dot1l 3′ UTR, but it was reduced in embryos co-injected with miR-216a.(C) Quantification of relative fluorescence in 1 dpf embryos injected with GFP reporter only or co-injected with miR-216a and the GFP reporter. Data represent means ± SEMs from 3 independent experiments. **p < 0.01 (Student’s t test).(D) One-dpf-old embryos injected at the 1-cell stage with 100 pg of GFP-dot1l 3′ UTR carrying mutations in all miR-216a MREs. Embryos were injected with the mutant reporter, either alone or co-injected with miR-216a.(E) Quantification of relative fluorescence in 1 dpf embryos injected with mutant GFP reporter alone or with co-injection of miR-216a. Data represent means ± SEMs from 3 independent experiments. **p < 0.01 (Student’s t test).Scale bars, 500 µm.

Fig 5: Dot1l Is Required for MG Dedifferentiation and Proliferation during Retina Regeneration(A) Control MO or dot1l MOs were injected and electroporated into the left eyes of Tg(1016tuba1a:gfp) zebrafish before intense light exposure (0 h). After 45 h, retinas were collected, sectioned, and immunostained using antibodies against GFP and PCNA. Nuclei were counterstained with TOPRO (blue).(B) Dot1l loss-of-function reduced tuba1a:GFP transgene expression and the number of INL PCNA+ proliferating cells.(C) Quantification of GFP+ dedifferentiated MG and PCNA+ proliferating progenitors in MO-ctl and MO-dot1l electroporated retinas. Data represent means ± SEMs, n = 5–6 fish; **p < 0.01 by 2-tailed Mann-Whitney U test.(D) Dose response to MO-dot1l. Increasing amounts of MO-dot1l MOs were injected and analyzed as in (B) and (C). Each data point represents an individual fish; *p < 0.05, **p < 0.01, and ***p < 0.001 by 1-way ANOVA.(E) Control vehicle (DMSO) or iDot1l (Dot1l inhibitor EPZ004777) were injected intravitreally into the left eyes of Tg(1016tuba1a:gfp) zebrafish before intense light exposure (0 h). After 45 h, retinas were collected, sectioned, and immunostained using antibodies against GFP and PCNA. Nuclei were counterstained with TOPRO (blue).(F) Quantification of total GFP+ and PCNA+ cells. Data represent means ± SEMs, n = 10 fish; *p = 0.0111; ****p < 0.0001 by 2-tailed Mann-Whitney U test.(G) Dose response to iDot1l. Increasing amounts of the Dot1l inhibitor were injected and analyzed as in (E) and (F). Each data point represents an individual fish; *p < 0.05, **p < 0.01, and ***p < 0.001 by 1-way ANOVA.GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer;. Scale bars, 50 µm.See also Figures S2 and S3.