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

First published online July 19, 2004
doi: 10.1242/10.1242/dev.01218

This Article Summary Full Text Full Text (PDF) 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 Otero, J. J. Articles by Kessler, J. A. Search for Related Content PubMed PubMed Citation Articles by Otero, J. J. Articles by Kessler, J. A.

ß-Catenin signaling is required for neural differentiation of embryonic stem cells

Department of Neurology, Northwestern University Feinberg School of Medicine, 303 East Chicago Avenue, Chicago, IL 60611, USA

View larger version (29K): [in a new window]   Fig. 4. Modulation of ß-catenin signaling regulates TCF/LEF signaling. ES cells were stably transfected with the constructs indicated (A). To test the activity of these constructs, we transiently transfected HEK293 cells with the constructs indicated and a TOPFLASH plasmid containing the TCF/LEF promoter driving a luciferase reporter gene (B). The full-length and N-terminal truncations were able to stimulate the TCF/LEF promoter but the C-terminal truncation, the armadillo truncation, and the N-plus C-terminal truncation constructs did not stimulate the promoter. TCF/LEF driven transcription was also increased by increasing the amount of ß-catenin transfected into the cells.

View larger version (60K): [in a new window]   Fig. 1. Phosphorylation of ß-catenin is inhibited in low-density cultures. (A) Western blot analyses of total cell lysates extracted on days 2 (d2) and 4 (d4) after ES cells were seeded for EB formation at low (LD, 105 cells/ml) and high densities (HD, 106 cells/ml). Density did not alter the levels of total ß-catenin protein. However, the levels of phospho-ß-catenin were significantly reduced in the low-density cultures when compared with the same time point in the high-density cultures using both antibodies (lower two blots). ES cells were seeded for EB formation at either low densities (B) or high densities (C) and stained with anti-ß-catenin antibody. Fluorescence intensity over a random cross-section of an EB demonstrated diffuse staining in low-density EBs (B'), whereas high-density EBs resulted in peaks and valleys in fluorescent intensities (C'). (D) The difference in fluorescence intensity (*P<0.05, **P<0.01) between the peaks and valleys was quantified and plotted graphically. HD EBs had a higher average difference between the peaks and the valleys, demonstrating that ß-catenin staining is more localized (B',C',D, y axis shows fluorescence intensity). (E) Undifferentiated ES cells were transiently transfected with an artificial TCF/LEF promoter and then seeded at high and low densities. Total luciferase activity was then assayed at day 1 post-seeding. High-density EBs repressed basal levels of TCF/LEF activity 2-fold (y axis shows relative luciferase units). (F) ES cells were differentiated by EB formation at low (105 cells/ml) and high (106 cells/ml) densities. On day 4, RT-PCR for Pitx2 was performed in cells treated with RA for 2 hours. Pitx2 is upregulated at low density by RA treatment but not at high density.

View larger version (78K): [in a new window]   Fig. 2. Neural differentiation of ES cells is regulated by cell density. ES cells were induced to differentiate by the 4-/4+ protocol at increasing seeding densities. After the induction, cells were plated onto PDL/laminin-coated coverslips and fixed 3-4 days post-plating. Coverslips were stained with anti-ß-tubulin 3 (red) and anti-nestin (green), and counterstained with Hoechst dye (blue) to visualize nuclei. (A) 105 cells/ml seeding density; (B) 106 cells/ml seeding density; (C) quantification of neuronal differentiation at different seeding densities (ND, none detected; error bars are s.d.). (D) Quantification of nestin-positive cells at low (105 cells/ml) and high (106 cells/ml) densities with or without RA treatment. Seeding at low density allows significant neural differentiation. Higher seeding density inhibits both neural and neuronal differentiation (LD, low density; HD, high density).

View larger version (58K): [in a new window]   Fig. 7. Overexpression of ß-catenin overcomes density-dependent inhibition of neural differentiation in an EB-independent protocol. ES cells transfected with the constructs indicated were induced to differentiate into the neural lineage in an EB-independent protocol (see Materials and methods) at a density of 106 cells/cm2. Cells were fixed and stained 8-10 days post-plating, with Hoechst dye (blue), and anti-nestin (green) and anti-ß-tubulin 3 (red) antibodies. (A) Empty vector transfected cells; (B) ß-catenin N transfected cells. Control cultures were devoid of nestin- and ß-tubulin 3-positive cells, whereas overexpression of ß-catenin N resulted in many nestin- and ß-tubulin 3-positive cells, suggesting that ß-catenin exerts both proneural and proneuronal effects. (C) Quantification of cells immunoreactive to antibodies specific to the antigens indicated in the key. (D) BrdU incorporation of ES cells. Left panel (plus Lif): undifferentiated ES cells stably transfected with either empty vector (black) or ß-catenin N (gray) were pulsed for 3 hours with BrdU and stained with anti-BrdU antibodies. No statistical difference was found. Right panel: ES cells stably transfected with either empty vector (black) or ß-catenin N (gray) were differentiated in vitro and pulsed with BrdU on day 6 post-Lif withdrawal. ß-catenin N expression resulted in decreased BrdU incorporation (*P<0.05; **P<0.01, by t-test; ND, none detected). Cells transfected with empty vector (E) or ß-catenin N (F) seeded at high density were analyzed for ß-catenin localization by optical sectioning. High-density cultures had a highly membrane localized stain for ß-catenin, whereas in the ß-catenin N overexpressor, the staining was more diffuse. To substantiate the visual observation, fluorescent intensities over a random cross section of the cells were plotted. In the control cells, there are clean peaks and valleys (E'), whereas in the ß-catenin N-transfected cells, the fluorescence is noisier (F').

View larger version (51K): [in a new window]   Fig. 3. Stimulation of endogenous ß-catenin signaling overcomes the inhibitory effects of high density on RA-treated cultures. Stimulation of endogenous signaling was mediated by treatment with Wnt3a-conditioned media and by overexpression of H2kd-E-cadherin. (A) HEK293 cells were transfected with TCF/LEF-luciferase and tested for luciferase activity after treatment with Wnt3a-conditioned media or after cotransfection of H2kd-E-cadherin with TCF/LEF-luc. H2kd-E-cadherin was able to increase luciferase activity, although Wnt3a treatment had a greater effect. (B) Changes in ß-catenin degradation were analyzed in ES cells after Wnt3a treatment. ES cells were induced by EB formation in high density and were either treated with control-conditioned media or with Wnt3a-conditioned media. Total cell lysates were extracted on day 2 of the differentiation (2 days Wnt3a treatment) and analyzed by western blotting (B). Treatment with Wnt3a-conditioned media reduced the amount of phospho-ß-catenin while not significantly altering the levels of total ß-catenin. Wnt3a treatment resulted in a decrease in the levels of phopshorylated ß-catenin signaling. ES cells were induced by EB formation at high density in either control-conditioned media (C) or Wnt3a-conditioned media (D), and differentiated by the 4-/4+ protocol. Wnt3a treatment resulted in many ß-tubulin 3-positive cells (D), whereas control media did not (C). Cells were stably transfected with an empty vector (E), H2kd-E-cadherin (F) or full-length ß-catenin (G), and induced by EB formation at high density using the 4-/4+ protocol. Overexpression of H2kd-E-cadherin and ß-catenin resulted in neuronal differentiation in high-density cultures, but transfection with empty vector did not. (C-G) All cells were fixed 3-4 days post-plating, and stained with anti-ß-tubulin 3 antibodies (green) and counterstained with Hoechst (blue).

View larger version (125K): [in a new window]   Fig. 5. The armadillo domain mediates neurogenesis promoted by ß-catenin and RA acts synergistically to enhance ß-catenin-mediated neurogenesis. Cells were seeded at high (106 cells/ml, HD) or low densities (105 cells/ml, LD) and treated with or without RA (cells in panels C,G,K,O were grown in KOSR media, see Materials and methods). Cells were fixed and stained 3-4 days post-plating with ß-tubulin 3 (green) and counterstained with Hoechst dye (blue). (A-D) Cells transfected with empty vector. (E-H) Cells transfected with ß-catenin Armadillo. (I-L) Cells transfected with ß-catenin C. (M-P) Cells transfected with ß-catenin N. The armadillo domain of ß-catenin, but not the C terminus, was required to overcome density-dependent inhibition of neurogenesis.

View larger version (12K): [in a new window]   Fig. 6. RA acts synergistically with ß-catenin to induce neuronal differentiation in ES cells. (A) The percentage of ß-tubulin 3-immunoreactive cells was quantified from high density inductions with or without RA treatment (see Materials and methods). ß-catenin N induced a greater proportion of neurons than ß-catenin C did. Treatment with RA, along with either ß-catenin C or N, resulted in an increase in the number of neurons, suggesting that there is synergy between RA and ß-catenin signaling (ND, none detected; error bars show s.d.). (B) Cells with neuronal morphology from these cultures were patch clamped to determine whether they expressed voltage-gated ion channels (see Materials and methods). All cultures showed voltage-gated ion channel activity. A representative tracing from neurons induced from ß-catenin C transfected ES cells at high density without RA is shown. Cells from control conditions did not exhibit this activity (data not shown). Cells were held at -90 mV holding potential and activated at -30 mV.

View larger version (91K): [in a new window]   Fig. 8. ß-catenin overexpression results in a less restricted population of neurons compared with RA-derived neurons. The phenotype of neurons generated by ß-catenin N and C overexpression in high-density EB cultures was compared with neurons generated by RA treatment of low-density EB cultures. (A-D) Hoxc4, red; ß-tubulin 3, green; (E-H) NeuN, green; GABA, red; (I-L) tyrosine hydroxylase, red; ß-tubulin 3, green; Hoechst stain blue (white arrows indicate TH-positive neurons); (M-P) Map2, green; ß-tubulin 3, red; (Q-T) synaptophysin, green; ß-tubulin 3, red. (A-D) All neurons generated by either ß-catenin overexpression or RA treatment were positive for Hoxc4, a homeobox gene specific for caudal neurons. (E-H) Many neurons were positive for the neurotransmitter GABA and all neurons were positive for NeuN. (I-L) Some neurons induced by overexpression of ß-catenin were immunoreactive for tyrosine hydroxylase (TH), the rate-limiting enzyme in catecholamine biosynthesis. By contrast, TH immunoreactivity was not observed in RA-treated conditions. (M-T) All ß-tubulin 3-immunoreactive cells also expressed Map2 and synaptophysin. (U) Quantification of the percentage of gabaergic neurons found in the cultures. There was no statistically significant difference between the RA-treated and untreated cultures (by ANOVA). (V) Neurons generated from untransfected ES cells differentiated at low density also express Hoxc4, suggesting that these are caudal neurons (green, Hoxc4; red, ß-tubulin 3).