QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online July 12, 2005
doi: 10.1242/10.1242/dev.01912

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fan, G. Articles by Sun, Y. E. Search for Related Content PubMed PubMed Citation Articles by Fan, G. Articles by Sun, Y. E.

DNA methylation controls the timing of astrogliogenesis through regulation of JAK-STAT signaling

¹ Department of Human Genetics, University of California at Los Angeles, 695 Charles Young Drive South, Los Angeles, CA 90095, USA
² UCLA MRRC, University of California at Los Angeles, 695 Charles Young Drive South, Los Angeles, CA 90095, USA
³ Departments of Molecular and Medical Pharmacology, and Psychiatry and Behavioral Sciences, University of California at Los Angeles, 695 Charles Young Drive South, Los Angeles, CA 90095, USA
⁴ UCLA Neuropsychiatric Institute, University of California at Los Angeles, 695 Charles Young Drive South, Los Angeles, CA 90095, USA
⁵ Department of Biological Chemistry, University of California at Los Angeles, 695 Charles Young Drive South, Los Angeles, CA 90095, USA
⁶ Department of Medicine, University of California at Los Angeles, 695 Charles Young Drive South, Los Angeles, CA 90095, USA

View larger version (73K): [in a new window]   Fig. 1. Precocious astroglial differentiation in Dnmt1–/– CNS in vivo. (A,B) Immunohistochemistry studies indicate enhanced S100ß- and GFAP-positive staining in E15.5 (A) and E18.5 (B) cervical spinal cords from littermate control (con) and Dnmt1–/–mutant (mut) mice. Bottom rows show enlargements of the boxed areas in the tops rows. (C,D) Western blot analysis of GFAP (C) and S100ß (D) proteins in E12-18 spinal cords at cervical/thoracic level. (E) GFAP immunostaining (red, arrows) of the E18.5 cortical VZ/SVZ in coronal sections (relatively caudal regions) (LV, lateral ventricle; D/M, dorsal/medial; V/L, ventral/lateral). (F) GFAP immunostaining of E15.5 hippocampal primordial areas. Scale bar: 157 µm. (G) Western blot analysis of GFAP protein in E18.5 brain samples. CTX, cortex; CB, cerebellum; TH, thalamus; STR, striatum; OB, olfactory bulb. Reblotting with an antibody against ßIII-tubulin serves as an internal control for loading. (H) Northern analysis of GFAP mRNA in E18.5 whole brain samples. con, control; mut, Dnmt1–/–.

View larger version (91K): [in a new window]   Fig. 2. Precocious astroglial differentiation in methylation-deficient E11.5 and E15.5 mouse CNS cultures. (A) Wild-type E11.5 mouse cortical precursor cells from Balb/c wild-type mice were dissociated and cultured for 2 days (2 d), 4 days (4 d) and 7 days (7 d) in the absence (con) and presence of LIF (50 ng/ml). Cells were stained with antibodies against a neuronal marker MAP2 (green) and GFAP (red). (B) E11.5 CNS cells from control (con) and Dnmt1–/– (mut) littermate embryos were cultured with or without LIF treatment for 2 days, and stained for GFAP (red) and a neural progenitor marker, nestin (green). Co-localization of nestin and GFAP (orange) in newly differentiated astrocytes in Dnmt1–/–mutant cultures (mut-LIF) indicates precocious astrocyte differentiation. (C) Western blot analysis of GFAP protein in two pairs of 3-day-old E11.5 NPC cultures. ß-actin serves as a sample loading control. (D) S100ß staining (red) of E11.5 CNS cells that were cultured for 4 days with LIF treatment in the last 2 days. DAPI nuclear counterstaining (blue) indicates similar cell densities between control (con) and Dnmt1–/– (mut) cultures. (E) 5'-methylcytosine (5'meC) antibody staining (red) of 3-day-old E11.5 control (con) and Dnmt1–/– (mut) NPCs. Counterstaining with DAPI (blue). The nuclear staining pattern is distinct, with heterochromatic punctuates intensely positive for 5'meC. (F) NPCs were infected with a GFP-expressing retrovirus on the first day of culturing. Two- to 3 days later, the virally infected GFP cells were stained for GFAP or MAP2, and the percentage of cells differentiating into either neurons or glia was measured and plotted (n=4). (G) Four-day cultured E15.5 cortical cells from control (littermate) and Dnmt1–/– mice in the presence and absence of LIF, were triple labeled with MAP2 (green), GFAP (red) and DAPI (blue). Scale bar: 32 µm.

View larger version (43K): [in a new window]   Fig. 3. Enhanced activation of JAK/STAT signaling in hypomethylated NPCs. (A) Western blot analysis of total STAT1 and pSTAT1 proteins in 1 day (1 D), 4 day (4 D) and 7 day (7 D) cultured E11.5 wild-type cortical cells with transient 20 minute LIF treatment. (B) Western blot analysis of total and phosphorylated STAT1/3 protein in 4-day-old cultured control (con) and Dnmt1–/– (mut) E11.5 CNS cells with 20 minutes LIF treatment. (C) Northern blot analysis of STAT1 and STAT3 mRNA in E18.5 CNS samples. (D) Western blot analysis of gp130 receptor protein in E18 cortices. con, control; mut, Dnmt1–/–.

View larger version (52K): [in a new window]   Fig. 4. Precocious astroglial differentiation is mediated by enhanced activation of JAK/STAT signaling in Dnmt1–/– NPCs. (A) Wild-type or a STAT-binding mutant form of the 1.9 kb rat GFAP promoter-luciferase reporter constructs were co-transfected with the renilla-TK control plasmid into 3 day cultured E11.5 control and Dnmt1–/– CNS NPC cultures. After 24 hours, cells were lysed and subjected to dual-luciferase assays (Promega). *P<0.001 compared with the control group (Dnmt1+/+) without LIF treatment. **P<0.01 compared with the group of Dnmt1–/– cells without LIF treatment (ANOVA with Post-hoc tests). (B) EMSA assay using a 25 bp unmethylated probe containing the STAT-binding element within the GFAP promoter with nuclear extracts from cultured control and Dnmt1–/– E11.5 CNS cells as in A. The identity of the DNA-protein complex (*) was characterized using anti-STAT1 and anti-STAT3 supershift assays (arrows). (C) Left panel, bFGF expanded (gliogenic) cortical progenitor cells were left untreated or treated with LIF for 30 minutes and subjected to chromatin immunoprecipitation (ChIP) assay with an antibody against STAT3 (Santa Cruz). A control antibody, anti-ß-galactosidase, was used to control for ChIP assay specificity. In the right two panels, ChIP assays were performed on 3-day-old cultured control (con) and Dnmt1–/– (mut) E11.5 CNS cells using the STAT1 and STAT3 antibodies. (D,E) E11.5 control Dnmt1+/+ and Dnmt1–/– CNS NPCs were cultured for 48 hours and co-transfected with a ß-gal-expressing construct and a control plasmid (con) or a dominant-negative STAT3F plasmid (STAT3F). After another 48 hours, cells were fixed and double-stained with antibodies against GFAP (green) and ß-gal (red), and counted for the percentage (mean±s.e.m.) of GFAP and ß-gal double-positive cells over total ß-gal positive cells. *P<0.01 compared with the Dnmt1+/+ (con) group. **P<0.01 compared with the group of Dnmt1–/– cells with ß-gal transfection (con) (ANOVA with Post-hoc tests).

View larger version (56K): [in a new window]   Fig. 5. Changes of DNA methylation on the GFAP and STAT1 promoters when NPCs become gliogenic in control and Dnmt1–/– cells. (A) Bisulfite sequencing analysis on eight CpG sites surrounding the STAT1/3 binding elements within the mouse Gfap promoter. The percentage of methylation at each of the 8 CpG sites was plotted. (B) Bisulfite sequencing analysis shows selective demethylation occurs at the –499 CpG site but not at the –594 CpG site during 24-96 hours of culturing period of wild-type E11.5 cortical cells. (C) Methylation-specific SNuPE assay was used to independently quantify the extent of methylation at the single CpG site lying within the STAT binding element in 1- and 4-day-old cultured E12.5 control (con) and Dnmt1–/– (mut) CNS cells and in E18.5 brain samples in vivo. (D) Bisulfite sequencing analysis of eight CpG sites within the Stat1 promoter (between –731 bp and –409 bp promoter region of the gene) in E18 control and Dnmt1–/– CNS samples. (E,F). E11 control (con) or Dnmt1–/– (mut) NPCs were transfected with either a ß-gal expression vector or a CAG-promoter-Dnmt1 expression plasmid (Chen et al., 2003) within the first 24 hours of cell culturing. After an additional 4 days of culturing in the presence of LIF (50 ng/ml) to promote glial differentiation, cells were double-labeled with GFAP/ß-gal or GFAP/Dnmt1 (E) and quantified for the percentage of GFAP+/ß-gal+ or GFAP+/Dnmt1+ cells as plotted in F. Dnmt1 overexpression cells can be easily detected by the strong Dnmt1 staining signals (arrows). Two arrowheads in the control culture indicate the typical nuclear staining pattern of the endogenous Dnmt1 protein. *P<0.001 compared with control (Dnmt1+/+) with ß-gal plasmid transfection. **P<0.001 compared with Dnmt1–/– cells with ß-gal plasmid transfection (ANOVA with Post-hoc tests).

View larger version (37K): [in a new window]   Fig. 6. Effect of DNA methylation on pSTAT association and activation of the Gfap, Stat1 and S100ß promoters. (A-E) ChIP assays to analyze the association of MeCP2 and STAT3 with the Gfap, Stat1 and S100ß promoters. (F-H) ChIP assays of histone H3 di-methyl lysine 9 (K9) and di- or tri-methyl lysine 4 (K4) within the Gfap, Stat1 and S100ß promoters. (I,J) E16 control and Dnmt1–/– cortical tissues were analyzed by ChIP assays for histone H3 di-methyl lysine 9 (K9) and di- or tri-methyl lysine 4 (K4) within the Gfap and Stat1 promoters

View larger version (27K): [in a new window]   Fig. 7. A model for DNA methylation-related glial gene chromatin remodeling during the switch from neurogenesis to gliogenesis. We have previously demonstrated that a positive-feedback loop for the JAK-STAT pathway allows for rapid activation of this pathway once it is derepressed (He et al., 2005). The time it takes to reach the threshold STAT activity for astroglial differentiation marks the onset of astrogliogenesis. DNA methylation serves as one of the key mechanisms blocking activation of the JAK-STAT pathway and glial cell lineage differentiation during the neurogenic period. Through a process of developmentally regulated DNA demethylation and active chromatin-remodeling, the JAK-STAT pathway is induced and astrocytic marker genes become responsive to STAT signaling, which marks the initiation of astrogliogenesis. In Dnmt1–/– NPCs, hypomethylation leads to accelerated activation of the JAK-STAT pathway, shortening the time required to reach the STAT activity threshold for astrocyte differentiation, leading to precocious astrogliogenesis.