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

First published online November 28, 2005
doi: 10.1242/10.1242/dev.02166

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 Bonaguidi, M. A. Articles by Kessler, J. A. Search for Related Content PubMed PubMed Citation Articles by Bonaguidi, M. A. Articles by Kessler, J. A.

LIF and BMP signaling generate separate and discrete types of GFAP-expressing cells

Davee Department of Neurology, Northwestern University's Feinberg School of Medicine, Chicago, IL 60611, USA

View larger version (84K): [in a new window]   Fig. 1. LIF and BMP both increase GFAP expression but promote different morphologies. (A) Western analysis demonstrates that endogenous BMP4 is present in E18.5 EGF-responsive neurosphere cultures. Preabsorption of the antibody with BMP4 (preabsorb) eliminates the band. (B,C) Quantitative PCR analysis illustrates that noggin inhibition of endogenous BMP decreases GFAP levels, whereas exogenous LIF and BMP4 increase GFAP levels in dissociated neurospheres plated for a 20-hour differentiation. Note that BMP treatment produces the highest level of GFAP. (D-F,H,I) Immunofluorescence reveals that 7-day LIF and BMP4 treatments promote different GFAP-expressing cell morphologies. Red, GFAP; blue insets, Hoechst nuclear counter stains. (D,H,J) BMP4 treatment produces a stellate morphology with an increased process number. (E,J) LIF treatment alone induces a mixture of elongated bipolar/tripolar and stellate morphologies. (F,I,J) In the presence of noggin, LIF promotes the bipolar/tripolar morphology and reduces process number. (G) Quantitation of GFAP immunofluorescent cells. Noggin treatment significantly reduces the number of GFAP+ cells. Conversely, LIF, and LIF plus noggin, treatment increases the number of GFAP+ cells and BMP4 treatment results in an even larger increase. (G) *P<0.05, **P<0.001; (J) *P<0.01, **P<0.001 ANOVA. Error bars are ±s.e.m. Scale bars: 20 µm in D-F;10 µm in H,I.

View larger version (55K): [in a new window]   Fig. 2. BMP4 prompts, whereas LIF reduces, exit of GFAP+ cells from the cell cycle. Neurosphere cells were plated and allowed to differentiate with or without cytokines for 6 days, followed by 16 hours of labeling with BrdU. (A,F) Only a small percentage of GFAP+ cells (green) incorporate BrdU (red). (B,F) Treatment with noggin to inhibit endogenous BMP signaling enhances BrdU incorporation. (C,D,F) Treatment with LIF alone does not significantly alter BrdU incorporation. However, LIF treatment in the presence of noggin greatly increases BrdU incorporation compared with treatment with noggin alone. (E,F) BMP4 treatment virtually abolishes BrdU incorporation. (F) Quantitation of the percentage of GFAP+ cells that incorporate BrdU. *P<0.01, **P<0.001 ANOVA. Scale bars: 20 µm in A-E.

View larger version (120K): [in a new window]   Fig. 3. LIF increases, whereas BMP4 decreases, neural precursor markers in GFAP+ cells. Neurosphere cells were plated and allowed to differentiate with or without cytokines for 7 days before immunocytochemical examination. (A,F) Few differentiated GFAP+ cells express the progenitor marker SOX1. (B-D,F) Treatment with LIF, noggin, and LIF plus noggin all increase the percentage of SOX1+GFAP+ precursors. (E,F) Conversely, BMP4 treatment decreases the percentage of SOX1+GFAP+ precursors. (G,L) Similarly, few differentiated GFAP+ cells express the neural stem cell marker LeX. (H,L) Treatment with noggin does not alter LeX expression on GFAP+ cells. (I,J,L) LIF treatment with or without noggin increases the number of LeX+GFAP+ cells. (K,L) By contrast, BMP4 treatment decreases the number of LeX+GFAP+ cells. Thus LIF and BMP have opposite effects on stem/progenitor cell markers in GFAP+ cells. *P<0.05, **P<0.005 ANOVA. Scale bars: 20 µm in A-F,H-K.

View larger version (80K): [in a new window]   Fig. 4. LIF promotes, whereas BMP4 inhibits, a GFAP-expressing multipotential stem cell fate and neuron production. (A) Experimental paradigm for the cytokine induction, rGFAP promoter and FACS selection, and stem cell analysis of GFAP+ cells. Neurospheres were expanded, infected with the rGFAPp-EGFP retrovirus, further expanded, plated for 5 days to differentiate with or without cytokines and then sorted on the basis of eGFP expression (green). Sorted cells were plated in a neurosphere-forming assay to assess self-renewal and subsequent spheres were plated for differentiation to assess the ability to form neurons (red), astrocytes (purple) and oligodendrocytes (blue). (B,G) In the neurosphere-forming assay, 2% of control GFAP+ cells self-renew (green indicates cells that are GFP positive). (C-E,G) LIF and noggin treatments each increase the percentage of GFAP+ cells that form neurospheres (green). LIF treatment also increases neurosphere size, which is further increased by additional BMP inhibition. (F,G) BMP4 treatment prevents the self-renewal of GFAP+ cells. (H) A single GFP-positive neurosphere plated for differentiation gives rise to oligodendrocytes (O4, blue), neurons (ß-tubulin, red) and astrocytes (GFAP, pink), demonstrating multipotentiality. (I-M) GFP-expressing neurosphere populations were dissociated to assess neuronal progeny. (I,M) GFAP-expressing (GFP positive, green) cells produce few neurons (red). (J,M) Noggin treatment alone does not significantly change neuron production. (K,M) LIF treatment significantly increases the number of neurons generated. (L,M) LIF treatment with BMP inhibition further increases neuron production. (G) *P<0.05, **P<0.01; (M) *P<0.01, **P<0.005 ANOVA. Scale bars: 100 µm in B-F; 20 µm in H-L.

View larger version (95K): [in a new window]   Fig. 5. LIF maintains GFAP-expressing multipotent progenitors. (A) LIF and low (1-2 ng/ml) EGF propagate neural progenitor cells as a monolayer for at least 10 passages. (B) At higher passages (p7 is shown), GFAP-expressing cells proliferate, as shown by BrdU incorporation (green, 2-hour pulse), and exhibit the progenitor marker LeX (red). (C) Higher passage cultures are still able to generate neurons (green), whereas nearby GFAP-expressing cells remain in cell cycle, as demonstrated by Ki67 expression (red). (D,E) To directly test the ability of GFAP-expressing cells for self-renewal and multipotentiality, higher passage cells were infected with the rGFAPp-EGFP retrovirus, selected by FACS, plated in a neurosphere-forming assay to assess self-renewal (in the absence of high mitogen), and subsequent spheres plated for differentiation to assess the ability to form neurons (red) and astrocytes (blue). (D) GFAP-expressing cells are able to self-renew at high passages, as shown by neurosphere formation (green, GFP). (E) Spheres formed from high passage GFAP-expressing cells retain the ability to produce neurons (red; blue indicates astrocytes; green, GFP). Scale bars: 20 µm in A-C,E; 100 µm in D.

View larger version (119K): [in a new window]   Fig. 6. BMPs regulate morphology and are necessary and sufficient for the cell-cycle exit of GFAP-expressing cells in vivo. Transgenic animals overexpressing BMP4 or noggin were analyzed at P15. (A-K) Immunofluorescence for GFAP and Ki67 in the hippocampus SGZ (A-D) and ML (E-K). (L) Plot of Ki67+GFAP+ cells in the SGZ (*P<0.03) and the ML (*P<0.01, **P<0.04). (M) Average number of processes per GFAP+ cell. Numbers in parentheses indicate the number of cells analyzed. *P<0.01, **P<0.001 ANOVA. (A,B,L) GFAP-expressing cells in the SGZ remain in cell cycle and are increased in number when noggin is overexpressed. Arrows indicate clusters of Ki67+ cells. (C,L) Overexpression of BMP4 in the SGZ promotes cell-cycle exit of GFAP+ cells. (D) High magnification confocal image demonstrating co-localization of Ki67 and GFAP in the SGZ of noggin mice. (F,G,L) Few GFAP-expressing cells in the ML remain in cell cycle in wild-type or BMP4-overexpressing mice. (E,L) BMP inhibition in the ML maintains GFAP+ cells in cell cycle. Arrows (E) indicate Ki67+GFAP+ cells. (H) High magnification confocal image demonstrating co-localization of Ki67 and GFAP in the ML of noggin mice. (J,K,M) GFAP+ cells in the ML exhibit branched morphologies that become further ramified with BMP4 overexpression. (I,M) Noggin overexpression in the ML prevents the formation of stellate morphology, and these cells resemble the thin elongated GFAP+ cells in the SGZ (H,D). GCL, granule cell layer; SGZ, subgranular zone; h, hilus; ML, molecular layer; WT, wild type; NN, NSE-Noggin; NB, NSE-BMP4. Scale bar in B: 20 µm for A-C,E-G; 10 µm for D,H-K.

View larger version (106K): [in a new window]   Fig. 7. BMPs regulate the maturation of GFAP-expressing cells in vivo. (A-H) GFAP (green), and SOX1 or vimentin (red) in the hippocampus SGZ (A-C,G) and ML (D-F,H) in P15 wild-type (B,E), NSE-noggin (A,D,G,H) and NSE-BMP4 (C,F) animals. Cells were counterstained with Hoechst (blue). (I,J) Number of SOX1-(I) and vimentin-(J) expressing GFAP+ cells. (A,B,D,I,J) GFAP-expressing cells in the SGZ remain as progenitors and increase in number when noggin is overexpressed, as assessed by SOX1 and vimentin co-labeling. (C,I,J) Overexpression of BMP4 in the SGZ promotes the loss of progenitor markers in GFAP+ cells. (D) Confocal image demonstrating co-localization of SOX1 and GFAP in the SGZ of noggin mice. (F,G,I,J) GFAP-expressing cells in the ML rarely co-express SOX1 or vimentin in wild-type or overexpressed BMP4 mice. (E,I,J) BMP inhibition in the ML increases progenitor markers in GFAP+ cells. (H) Confocal image demonstrating co-localization of SOX1 and GFAP in the ML of noggin mice. (I) *P<0.02, **P<0.05; (J) *P<0.03, **P<0.01 ANOVA. Scale bar: in C, 10 µm for A-C,G,H; in E, 20 µm for D-F.

View larger version (20K): [in a new window]   Fig. 8. Model for the affects of LIF and BMP on astrocyte differentiation. Stem cells in the developing rodent brain are initially GFAP-, but later express GFAP during the postnatal and adult periods. LIF fosters maintenance of stem/progenitor trait throughout the continuum, but converts GFAP- cells into GFAP+ ones. By contrast, BMPs induce exit from cell cycle and the loss of stem/progenitor cell traits, resulting in what we have termed a mature astrocyte. Inhibition of endogenous BMPs is required to prevent maturation of adult stem cells into the mature astrocytic phenotype.