MicroRNA-mediated control of oligodendrocyte differentiation

Abstract

MicroRNAs (miRNAs) regulate various biological processes, but evidence for miRNAs that control the differentiation program of specific neural cell types has been elusive. To determine the role of miRNAs in the formation of myelinating oligodendrocytes, we selectively deleted a miRNA-processing enzyme, Dicer1, in oligodendrocyte lineage cells. Mice lacking Dicer1 display severe myelinating deficits despite an expansion of the oligodendrocyte progenitor pool. To search for miRNAs responsible for the induction of oligodendrocyte maturation, we identified miR-219 and miR-338 as oligodendrocyte-specific miRNAs in spinal cord. Overexpression of these miRNAs is sufficient to promote oligodendrocyte differentiation. Additionally, blockage of these miRNA activities in oligodendrocyte precursor culture and knockdown of miR-219 in zebrafish inhibit oligodendrocyte maturation. miR-219 and miR-338 function in part by directly repressing negative regulators of oligodendrocyte differentiation, including transcription factors Sox6 and Hes5. These findings illustrate that miRNAs are important regulators of oligodendrocyte differentiation, providing new targets for myelin repair.

Figure 1. Dicer1 is Required for Oligodendrocyte Myelination

A-B) Immunocytochemistry using antibodies to mature oligodendrocyte makers (MBP and CC1), an astrocyte marker GFAP and in situ hybridization using probes to Plp1 on frozen sections of P14 spinal cord and brains from control (Ctrl, Olig1Cre+/-;Dicer1^lox/+) and Dicer1CKO (Olig1Cre+/-;Dicer1^lox/lox) mice. Expression of MBP, CC1 and Plp1 is evident in white matter (arrows) of Ctrl but not Dicer1CKO mice. C) Electron micrographs of spinal cord and optic nerves from Ctrl and Dicer1CKO mice at P14. Multilamellar myelin sheaths are apparent around many axons in Ctrl mice and axons in Dicer1CKO mice are essentially unmyelinated (red arrow). D) Histogram depicts quantification of myelinated axons in the optic nerve and spinal cord of Ctrl and Dicer1CKO mice per defined area (100 μm²), respectively. Error bars indicated mean± S.D. * P < 0.01, two-tailed paired Student's t test. E) Histogram shows the qRT-PCR analysis of Dicer1 expression from RNAs isolated from intact optic nerves of Dicer1^lox/+;Olig1Cre (Ctrl) and Dicer1CKO mice (n=3) at p14 with primers against the floxed exon at the Dicer1 locus. GAPDH was used as internal control. *P<0.01. Scale bars in A and B: 200 μm; in C, 5 μm for upper panel and 2 μm for lower panel.

Figure 2. Oligodendrocyte Precursor Proliferation in the Cortex of Dicer1 Mutants

A-C) Immunocytochemistry using antibodies to Olig2, BrdU and PDGFRα on frozen sections of P7 (B) and P14 (A-C) brains from Olig1Cre;Dicer1^lox/+ (Ctrl) and Olig1Cre;Dicer1^lox/lox (Dicer1CKO) mice. D) Immunocytochemistry using antibodies to BrdU and PDGFRα on frozen sections of P14 brains from Olig1CreDicer1^lox/+ (Ctrl) mice. Scale bars: 100 μm. Arrows in B-D indicate co-labeling cells. E-F) Histograms depict the quantification of Olig2 and BrdU expressing cells. Error bars indicated mean± S.D. * P < 0.001. G) Histogram depicts the percentage of BrdU expressing cells among Olig2 expressing cells. Error bars indicated mean± S.D. * P < 0.001.

Figure 3. Identification of miR-219 and miR-338 as Oligodendrocyte-Specific miRNAs in the Spinal Cord

A-C) miRNAs isolated from spinal cord and optic nerve tissues from WT, Dicer1CKO or Olig1 null mice were subjected to miRNA microarray analysis. The log₂ transformation of relative fold changes in the miRNA expression level from WT versus Dicer1CKO or Olig1KO mice are indicated. D) Expression of miR-219 and miR-338 in WT and Olig1 null spinal cord was analyzed by Northern blot. U6 RNA is used as a loading control. E-F) miR-219 (E) and miR-338 (F) expression was analyzed by qRT-PCR from WT mouse spinal cords at stages indicated. U6 RNA was used as internal control. G) Expression of miR-219 and miR-338 was examined in the spinal cord of WT and Olig1KO mice at P14 by in situ hybridization as indicated. Arrows indicate the miRNA expressing cells in the white matter region (left part of dashed lines); arrowheads indicate the gray matter. Left panels show at lower magnification, while middle and right panels show at higher magnification. H-I) Expression of miR-219 and miR-338 was examined in spinal cord of P14 WT mice by in situ hybridization, followed by immunostaining with CC1 and Olig2. Arrows indicate colabeling of miR-219 or miR-338 with oligodendrocyte markers CC1 and Olig2 in the same cells. Scale bars, in G, 100 μm; in H and I, 50 μm.

Figure 4. Enforced miR-219 and miR-338 Expression Promotes Oligodendrocyte Maturation in vitro

A-B) Mouse oligodendrocyte precursor-enriched cultures were transfected with miR-219, miR-338(5p+3p) mimics and scrambled miR control as indicated. The precursor cells were cultured in oligodendrocyte growth medium for another 4 days and subjected to immunostaining with PDGFRα (blue), RIP (red) and MBP (green). Scale bars: 50 μm. C-E) Histograms depict the percentage of RIP+ (C), MBP+ (D) and PDGFRα+ (E) cells among Olig2+ oligodendrocyte lineage cells, which represents all stages of oligodendrocytes. Data represent mean ± S.D. from three independent experiments. * P < 0.01. F) Oligodendrocyte precursor-enriched cultures from Dicer1CKO mice were transfected with miR-219, miR-338 (5p+3p) mimics and scrambled miR control as indicated. Cells were cultured in oligodendrocyte differentiation medium without mitogens for 3 days and subjected to immunostaining with RIP (red) and MBP (green). Scale bars: 50 μm. G) Histogram depicts the number of MBP+ cells per defined area (0.14 mm²). Data represent mean ± S.D. from three independent experiments. * P < 0.01.

Figure 5. Ectopic Expression of miR-219 and miR-338 Promotes Precocious Oligodendrocyte Differentiation in Developing Chick Neural Tube and Embryonic Mouse Cortex

A-B) Chick neural tubes were electroporated with pCIG expression vectors for miR-219 (A) and miR-338 (B) into the neural tube at E2.5 and harvested at E5.5. The spinal cord sections were analyzed by in situ hybridization with probes to Pdgfrα and by immunostaining with antibody against Olig2 or MBP. GFP expression indicates the electroporated side. Arrows indicate ectopic expression of Pdgfrα, Olig2 and MBP detected on the electroporated side of chick neural tubes, respectively. Scale bars, 100μm. C) Schematic diagram depicts in utero electroporation in the mouse developing cortex at e14.5. Electroporated embryonic cortices (n =5) were harvested and analyzed at e17.5, at which stage there is absent of endogenous MBP expression. D) Quantification of the number of Olig2+ or PDGFRα+ cells on the electroporated cortices at a defined cortical area (0.1 mm²). Y axis indicated the ratio of the number of Olig2+ or PDGFRα+ cells in miR-219 or miR-338 overexpressing cortices to that in the control. Error bars indicated mean± S.D. * P < 0.05. E) Mouse embryos at e14.5 were electroporated with expression vectors for miR-219, miR-338 and RFP, respectively, and harvested at e17.5. The sections of electroporated cortices were analyzed by immunostaining with antibodies to MBP. Arrows indicate the electroporated cells expressing MBP. F) Triple immunostaining revealed that miR-219 or miR-338 transfected cells express MBP and Olig2 in the electroporated cortices. Arrows indicate co-labeling cells. Scale bars, in E, 100μm; in F, 25 μm.

Figure 6. Inhibition of miR-219 and miR-338 Blocks Oligodendrocyte Maturation

A) Purified rat oligodendrocyte precursor cells were transfected with LNA knockdown antisense miRNAs and scrambled control oligonucleotides as indicated and cultured in oligodendrocyte differentiation medium. Four days after transfection, cultures were subjected to MBP immunostaining. B) Histogram depicts the relative number of MBP+ cells in per defined area (0.14 mm²) treated with LNA-anti-219 and LNA-anti-338(5p+3p) compared with that treated with scrambled control oligonucleotide. Data represent mean ± S.D. from three independent experiments, *P<0.01. C) Expression of miR-219 in WT and miR-219MO injected larvae was analyzed by Northern blot. U6 RNA is used as loading control. D-E) Stereomicroscope images showing a lateral view of living Tg(olig2:egfp) zebrafish larvae at 3 dpf with or without morpholino miR-219 injection. E is shown at larger magnification to show the OPCs in the dorsal spinal cord in control larvae but their absence in the miR-219 MO-injected larvae as indicated by arrows. F) Transverse sections through the spinal cord of larvae at 3 dpf were immunostained with antibody to Sox10. Sox10+ OPCs are indicated by arrows. G) Quantification of the average number of Sox10+ OPCs per transverse section in miR-219MO injected and control zebrafish (*P<0.01). Error bars indicated mean± SD. H) qRT-PCR analysis of expression of mbp and ef1a encoding eukaryotic elongation factor 1 from RNAs isolated from miR-219 MO injected and control zebrafishes. rpl13 encoding a ribosomal protein was used as an internal control. *P<0.01. I-J) In situ hybridization with an mbp probe in zebrafish embryos with or without miR-219MO injection. Arrows in I (lateral view) indicate the formation of mature oligodendrocytes in the dorsal spinal cord. Brackets indicate mbp expression in ventral spinal cord. Arrows in J (cross-section) indicate mbp expression. Scale bars in A, 50μm; C-D40 μm; E, 20 μm, G, 40 μm.

Figure 7. miR-219 and miR-338 Target Oligodendrocyte Differentiation Inhibitors

A) Sequence analysis of 3′UTR of mouse Sox6 and Hes5 transcripts. miR-219, miR-338-5p and miR-338-3p recognition sites are indicated by red, green and blue bars, respectively. Black lines underneath depict the regions (UTR-I or UTR-II) of 3′ UTRs that cloned into pMIR-REPORTER: Sox6 3′UTR-I, 863bp; Sox6 3′UTR-II, 1380bp; Hes5 3′UTR, 630bp. B-C) Luciferase reporter assays for the effects of expression of miR-219, miR-338-3p or -5p and their mutant forms with corresponding “seed” sequence mutations on activities of reporters carrying the 3′UTR segments of Sox6 or Hes5. The histogram shows the ratio of the luciferase activity normalized to control expression vector carrying the reverse sequence of miR-219 and miR-338-3p and miR-338-5p, respectively. Data represent mean ± SDs from three independent experiments. * P < 0.01. D-E) Luciferase reporter assays for the effects of expression of miR-219 and miR-338 on activities of reporters carrying WT and mutant 3′UTR of Sox6 and Hes5. Sox6-3′UTR-IMu and Sox6-3′UTR-IIMu represent Sox6-3′UTR reporters with mutations of a miR-219 binding site in UTR-I and two miR-338-5p binding sites in UTR-II, respectively, as shown in A. The histogram shows the ratio of the luciferase activity normalized to control expression vectors carrying miR-219rev and miR-338rev, respectively. Data represent mean ± SDs from three independent experiments. * P < 0.01. F) qRT-PCR analysis of Sox6 expression from RNAs isolated from mouse primary OPC cultures 4 days after transfection with miR-219, miR-338 (5p+3p) mimics or both. Scrambled miR transfection was included as control. * P < 0.01. G) Western blot analysis shows that Sox6 protein was downregulated by miR-219 or miR-338 overexpression compared with vector-transfected cells. GAPDH was included as an internal control. H) qRT-PCR analysis of Hes5 expression from RNAs isolated from mouse primary OPC cultures transfected with miR-219, miR-338 (5p+3p) mimics or both. Scrambled miR transfection was included as control. * P < 0.01. I-J) Expression of Sox6 and Hes5 in rat hippocampal neural precursor cells was examined by immunostaining after 3 days transfection with miR-219 or miR-338-expressing and control vectors. Arrows indicate the transfected cells. Scale bars: 50μm. K-L) Quantification of Sox6 (K) and Hes5 (L) positive cells among transfected cells (GFP+) from three independent above experiments. * P < 0.01. M-N) qRT-PCR analyses of Sox6 (M) and Hes5 (N) expression using RNAs isolated from mouse primary OPC cultures 4 days after transfection with LNA-miR-219, LNA-miR-338 (5p+3p) or both knockdown probes as well as the scrambled oligonucleotide control (ctrl) as indicated. Histograms indicated that knockdown of these miRNAs led to significant upregulation of Sox6 and Hes5. * P < 0.01. GAPDH was included as an internal control.

Figure 8. Schematic Diagram for the Potential Function of miR-219 and miR-338 in Regulating Oligodendrocyte Differentiation

A) In oligodendrocyte precursor cells (OPCs), mature miRNAs (e.g. miR-219 and miR-338) and their processing enzyme Dicer1 are present at a relatively low level. The amount of miRNAs in OPCs is not sufficient to repress the expression of oligodendrocyte differentiation inhibitors (e.g. Sox6 and Hes5) and/or pro-neuronal differentiation factors (e.g. NeuroD1 and Zfp238), which thereby prevent OPC differentiation. B) During oligodendrocyte differentiation, expression of mature miRNAs, particularly miR-219 and miR-338, and Dicer1 increases substantially. miR-219 and miR-338 permit or activate pro-OL factors for oligodendrocyte differentiation and maturation by downregulating a cohort of oligodendrocyte differentiation inhibitors, and repressing genes involved in neuronal differentiation. An important future study would be to identify the molecules that regulate temporal expression of these miRNAs during oligodendrocyte lineage progression. OPCs, oligodendrocyte precursor cells. OL, oligodendrocytes.