miR-132 Downregulation Causes Glial and Spinal Cord Maturation Deficits

(A) In situ hybridization of miR-132 at 48 hpf. Expression of miR-132 in the spinal cord (whole mount) and in the radial glial cells (cross-section). Arrowhead indicates region of cross-sectioning. Scale bars, 1 mm (lateral view); 4 µm (cross-section).

(B) Semiquantitative real-time PCR analysis of miR-132 expression in wild-type embryos and larvae at 10, 24, 48, and 72 hpf. Values were normalized to the U6 small nuclear RNA.

(C) miR-132 knockdown using an antisense morpholino. Top: Bright-field images of UIC, larvae injected with a 5b-MM or a morpholino against miR-132 (miR-132 MO) at 72 hpf, and schematic illustration of a zebrafish larva indicating the trunk region between S10 and S15 used to score the phenotypes. Scale bar, 1 mm. Middle: Lateral views of the trunk. The Gfap⁺ glial processes are immunostained with zrf-1. White arrowhead indicates intraprocess ectopic branches. Red arrowhead indicates ectopic branch connecting two neighboring processes. Schematic illustration depicts a sagittal view of the perineural glial processes. Scale bar, 50 µm. Bottom: Cross-sections of the trunk immunostained with zrf-1. Arrow indicates the spinal cord. Arrowheads indicate ectopic branches. Schematic illustration depicts a transverse section showing the normal arborization pattern of glial processes. Scale bar, 10 µm.

(D) Transverse sections of the spinal cord of Tg(gfap:GFP) embryos after BrdU incorporation at 48 hpf, quantified in (G) and (H). Scale bar, 4 µm.

(E) Quantification of the Gfap⁺ glial arborization defects in miR-132 morphants. Sample sizes: n = 172 for UIC; n = 132 for 5b-MM; and n = 195 for miR-132 MO.

(F) Lateral view of the spinal cord of miR-132 morphant and control Tg(olig2:GFP) embryos at 48 hpf, quantified in (I). Scale bar, 10 µm.

(G and H) Quantification of proliferating glial progenitors and of total number of progenitors. Values were expressed as percentage of BrdU-positive (BrdU⁺) cells to total number of Gfap⁺ cells within the spinal cord (G) or as total number of Gfap⁺ cells (H).

(I) Quantification of the total number of oligodendrocyte lineage precursor cells at 48 hpf.

(J) Spinal cord cross-sections of miR-132 morphants (miR-132 MO) and controls immunostained with zrf-1 depicting the primitive lumen (dashed rectangle, PL) at 24 hpf or the central canal (arrowheads, CC) at 72 hpf. Scale bars, 4 µm.

(K and L) Quantification of the diameter of the primitive lumen (K) or the central canal (L) in miR-132 morphants and controls.

In (G) through (I), (K), and (L) each dot of the scatterplot represents the normalized average from four serial optical sections (z stacks) of one spinal cord cross-section per embryo. Values are presented as mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ns, nonsignificant.

See also Figures S1 and S2.

The miR-132 Morphant Phenotype Is Mediated through the Transcriptional Corepressor Ctbp2

(A) Semiquantitative real-time PCR analysis showing the effect of Ctbp2 downregulation on her4 levels (CTBP2 MO, splice blocking morpholino) in miR-132 morphant and control embryos at 24 hpf. Values were normalized to the geometric mean of ef1a, b2m, and actin and expressed as fold change compared to the control.

(B and C) Effect of Ctbp2 downregulation using a splice blocking morpholino on the total number of Gfap⁺ (B) and Olig2⁺ (C) in the spinal cord of Ctbp2 morphant and control Tg(gfap:GFP) (B) or Tg(olig2:GFP) (C) embryos at 48 hpf. Each dot of the scatterplot represents the normalized average from four serial optical sections (z stacks) of one spinal cord cross-section per embryo.

(D) Transverse trunk sections of Tg(gfap:GFP) larvae either UIC, overexpressing gfap:mCherry (mCherry under the gfap promoter), or gfap:ctbp2 (ctbp2 under the gfap promoter) in a mixture with gfap:mCherry at 72 hpf. Quantified in (F). Scale bar, 10 µm. Arrowheads indicate aberrant arborization of glial processes.

(E) Zrf-1 immunostaining of whole mounts at 72 hpf, lateral views of trunk. StdCtr MO, standard control morpholino; CTBP2 MO1, translation blocking morpholino against Ctbp2; CTBP2 MO2, splice blocking morpholino against Ctbp2. Quantified in (G). Scale bar, 50 µm. Arrowheads indicate aberrant arborization of glial processes.

(F) Quantification of glial arborization phenotype at 72 hpf in gfap:Ctbp2-overexpressing larvae and control larvae overexpressing gfap:mCherry. Sample size, n = 75 for gfap:Ctbp2; n = 71 for gfap:mCherry.

(G) Quantification of glial arborization phenotype in miR132/Ctbp2 double morphants at 72 hpf. Sample sizes: n = 162 for UIC; n = 186 for miR-132 MO; n = 128 for CTBP2 MO1; n = 86 for miR-132 MO-CTBP2 MO1; n = 118 for CTBP2 MO2; n = 100 for miR-132 MO/CTBP2 MO2; n = 125 for 5b-MM; n = 70 for 5b-MM/CTBP2 MO1; n = 61 for 5b-MM/CTBP2 MO2; n = 120 for StdCtr; and n = 103 for miR-132 MO/StdCtr.

(H) Western blot analysis of Ctbp2 protein levels in miR-132 morphants compared to control groups at 72 hpf. Protein levels were normalized against Gapdh.

Values are presented as mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ns, nonsignificant. See also Figure S4.

miR-132 Regulates Ctbp2 Levels by Directly Binding to Its 3′ UTR

(A) Schematic illustration of the miR-132 MREs in the 3′ UTR of Ctbp2 in zebrafish (Danio rerio) and human (Homo sapiens) showing the three sites selected for the targeting by target protector morpholinos (TPs).

(B) Semiquantitative PCR analysis of the effect of the TPs on her4 expression at 72 hpf. Values were normalized to the geometric mean of ef1a, b2m, and actin and expressed as fold change compared to the control.

(C and D) Effect of the TPs on the total number of Gfap⁺ (C) and Olig2⁺ (D) in the spinal cord of morphant and control Tg(gfap:GFP) (C) or Tg(olig2:GFP) (D) embryos at 48 hpf. Each dot of the scatterplot represents the normalized average from four serial optical sections (z stacks) of one spinal cord cross-section per embryo. Values were compared to the control-injected group.

(E) Zrf-1 immunostaining of whole mounts at 72 hpf, lateral views of trunk. Arrowheads indicate ectopic branches of Gfap⁺ glial processes. TP, target protector morphants; TP/CTBP2 MO, target protector morphants coinjected with a translation blocking morpholino against Ctbp2; TP/Std Ctr, target protector morphants coinjected with a standard control morpholino. Scale bar, 50 µm.

(F) Quantification of glial arborization phenotype at 72 hpf. Sample sizes: n = 132 for TP1; n = 90 for TP1/CTBP2 MO; n = 104 for TP1/Std Ctr; n = 142 for TP2; n = 116 for TP2/CTBP2 MO; n = 107 for TP2/Std Ctr; n = 128 for TP3; n = 80 for TP3/CTBP2 MO; and n = 72 for TP3/Std Ctr.

Values are presented as mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ns, nonsignificant.

The Histone Deacetylase Sirt1 Is Regulated by miR-132 and Ctbp2

(A) Model of Sirt1 involvement in the miR-132/Ctbp2/Notch regulatory pathway.

(B) Semiquantitative PCR analysis of her4 expression levels in Sirt1 morphants (SIRT1 MO) and control embryos (Std Ctr) at 24 hpf. Values were normalized as described earlier.

(C and D) Effect of Sirt1 downregulation (SIRT1 MO) on the total number of Gfap⁺ (C) and Olig2⁺ (D) in the spinal cord of Sirt1 morphant and control Tg(gfap:GFP) (C) or Tg(olig2:GFP) (D) embryos at 48 hpf. Each dot of the scatterplot represents normalized averages as described earlier.

(E) Whole-mount immunostaining with zrf-1 at 72 hpf, lateral views of trunk. SIRT1 MO, Sirt1 morphants; StdCtr MO, larvae injected with standard control morpholino. Scale bar, 50 µm.

(F) Quantification of glial arborization phenotype in Sirt1 morphants at 72 hpf. Sample size, n = 75 for UIC; n = 90 for SIRT1 MO; n = 95 for StdCtr MO.

(G) Lateral trunk views of 72 hpf - larvae injected with miR-132 morpholino (miR-132 MO), gfap:mCherry (mCherry under the gfap promoter), gfap:sirt1 (sirt1 under the gfap promoter) or coinjected with miR132 morpholino and gfap:mCHERRY or miR-132 morpholino and gfap:sirt1. Scale bar, 50 µm. In (E) and (G), arrowheads indicate ectopic Gfap⁺ glial processes.

(H) Quantification of the partial rescue of the glial arborization phenotype in the miR-132 morphants overexpressing Sirt1 under the gfap promoter at 72 hpf. Sample sizes: n = 52 for miR-132 MO; n = 61 for gfap:mCherry; n = 57 for miR-132 MO/gfap:mCherry; n = 63 for gfap:sirt1; n = 72 for miR-132 MO/gfap:sirt1; n = 70 for 5b-MM; and n = 58 for 5b-MM/gfap:sirt1.

(I and J) Semiquantitative real-time PCR analysis showing expression of sirt1 in miR-132 morphant (miR-132 MO) and control embryos (5b-MM) at 24 hpf (I) or in gfap:GFP⁺ FACS-sorted cells isolated from Tg(gfap:GFP) Ctbp2 morphants compared to control larvae, at 72 hpf (J). Values were normalized as described earlier.

(K) Quantification of glial arborization phenotype in Ctbp2/Sirt1 double morphants. Sample sizes: n = 78 for CTBP2 MO (translation blocking morpholino); n = 95 for CTBP2 MO/SIRT1 MO; and n = 87 for SIRT1 MO.

(L) Quantification of the glial arborization phenotype at 72 hpf in UIC, Sirt1 morphants (SIRT1 MO), and larvae injected with a standard control morpholino (StdCtr) following 48 hr incubation in DAPT or vehicle control (DMSO). Sample sizes: n = 87 for UIC-DAPT; n = 74 for UIC-DMSO; n = 66 for SIRT1 MO-DAPT; n = 69 for SIRT1 MO-DMSO; n = 54 for StdCtr-DAPT; and n = 66 for StdCtr-DMSO. Values are presented as mean ± SEM. *p < 0.05; ***p < 0.001; ns, nonsignificant.