| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TISSUE-SPECIFIC STEM CELLS |
Department of Anatomy and Neurology, Neuroscience Training Program, and Waisman Center, University of Wisconsin-Madison, Madison, Wisconsin, USA
Key Words. Epidermal growth factor • Neurogenesis • Cortical progenitor • Stem cell • Cerebral cortex • Development
Correspondence: Correspondence: Clive N. Svendsen, Ph.D., Waisman Center, University of Wisconsin-Madison, 1500 Highland Avenue, Madison, Wisconsin 53705-2280, USA. Telephone: 608-265-8668; Fax: 608-263-5267; e-mail: svendsen{at}waisman.wisc.edu
Received on April 23, 2007;
accepted for publication on November 13, 2007.
First published online in STEM CELLS EXPRESS November 21, 2007.
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Disclosure of potential conflicts of interest is found at the end of this article.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
We have also shown that glutamate and a neurosteroid, dehydroepiandrosterone (DHEA), can significantly increase the proliferation rates of hNPC, and they mainly increase the elongated cells possessing morphological characteristics like those of radial glial cells [8, 9]. Radial glial cells are classically defined as cells with a radial morphology and glial characteristics typical of astrocytes, such as glial fibrillary acidic protein (GFAP) expression in primate [10]. In the developing brain, radial glial cells have long been known to produce cortical astrocytes, but recent data indicate that radial glial cells also possess neurogenic potential and divide asymmetrically to produce cortical neurons that migrate along their own processes to the appropriate cortical layers [11–13]. Similar radial glial cells have also been found in the human fetal cortex, although no real-time examination of their division and migration has been possible [14–16]. In some cases, these neurogenic radial glial cells isolated from rodents may also be stimulated to divide in vitro within cultures derived from the developing cortex [17]. However, the exact composition of neurospheres grown in culture from the developing human cortex is currently only poorly understood.
Furthermore, there have been few systematic studies to determine how different concentrations of EGF may affect the growth and differentiation of hNPC in vitro. Interestingly, human embryonic stem cells exposed to higher concentrations of FGF-2 showed increased proliferation and maintenance of pluripotency [18]. In this study, we established that high EGF concentrations, raised from 20 to 100 ng/ml in the culture medium, increased overall growth rates of hNPC. This was associated with decreased cell death, possibly because of activation of ErbB2 activity. We also found that increasing the concentration of EGF in hNPC neurosphere cultures led to increased numbers of cells with a "radial glial"-like morphology within neurospheres and a subsequent increase in the rate of neurogenesis.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Human cortical progenitors were prepared from fetal brains and induced to proliferate as neurospheres using established passaging methods to achieve optimal cellular expansion as previously described in detail [5]. Briefly, freshly dissected fetal tissue was dissociated in trypsin and seeded into flasks (75 cm2) at a density of 200,000 cells per milliliter of maintenance medium (Dulbecco's modified Eagle's medium/Ham's F-12 medium [7:3] containing penicillin/streptomycin/amphotericin B [1% vol/vol]) supplemented with B27 (2% vol/vol; Invitrogen, Carlsbad, CA, http://www.invitrogen.com), EGF (20 ng/ml; Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com), and FGF-2 (20 ng/ml; R&D Systems Inc., Minneapolis, http://www.rndsystems.com) with heparin (5 µg/ml; Sigma-Aldrich). Neurospheres were passaged using a chopping method using a McIlwain tissue chopper (Campden Instruments Ltd., Loughborough, U.K., http://www.campden-inst.com) [5]. Neurospheres were not allowed to grow any larger than 500 µm in diameter and were sectioned into 200-µm-diameter spheres. All cultures were maintained in a humidified incubator (37°C, 5% CO2 in air), and half the growth medium was replenished every 3–4 days. At 2 weeks after the first passage, cells were switched to maintenance medium containing N2 supplement (1%; Invitrogen) and 20 ng/ml EGF. After 10 weeks of growth, 10 ng/ml LIF (Chemicon, Temecula, CA, http://www.chemicon.com) was added to the cultures and maintained for a minimum of 4 weeks. Three independent lines of hNPC derived from three cortex samples (M031, M046, and M067) were used for this study. hNPC were maintained in the presence of either 20 or 100 ng/ml EGF and 10 ng/ml LIF (Chemicon) for a minimum of 14 days prior to experimentation. The concentrations of EGF protein in the conditioned medium were measured by enzyme immunoassays (human EGF Quantikine immunoassay kit; R&D Systems).
Growth and Proliferation Measurements
Single neurospheres from either 100 or 20 ng/ml EGF cultures were placed in a 96-well plate and were measured via Integrated Morphometry Analysis using Metamorph software (Molecular Devices, Dowington, PA, http://www.moleculardevices.com). Sphere volume was calculated on day 0 and every third day up to 15 days after plating. Half of the medium was exchanged every third day. Results are plotted as the log of the sphere volume, and 95% confidence intervals were calculated and are shown around the linear regression line (Fig. 1A).
|
Differentiation of hNPC Cultures
To determine the phenotypes of cells within neurospheres as well as measure the progenitors' innate ability to produce neurons and astrocytes, whole neurospheres from both high- and low-EGF cultures were dissociated and plated onto laminin-coated substrates in the absence of exogenous growth factors. Cells were allowed to plate for 2 hours (acute) or 7 days (differentiation) and were fixed using 4% paraformaldehyde (PFA). Cells were then processed for immunocytochemistry for β-tubulin-III (TuJ1; neurons) or GFAP (astrocytes).
Immunocytochemistry
Cells were processed for immunocytochemistry by fixing in either ice-cold methanol (for BrdU) or 4% PFA (for TuJ1, GFAP, or Nestin) and staining as described previously [8, 9]. Cells were permeabilized in a 5% normal goat/5% normal donkey serum with phosphate-buffered saline (PBS) and 0.2% Triton X-100 for 35 minutes at room temperature. After rinsing with PBS, the cells were then incubated with primary antibodies against human GFAP (1:1,000, mouse monoclonal IgG1; Chemicon), human Nestin (1:3,000, mouse monoclonal IgG1; Chemicon), TuJ1 (1:3,000, mouse monoclonal IgG2b; Sigma-Aldrich), and brain lipid-binding protein (BLBP; 1:2,000, rabbit polyclonal; kindly donated by Dr. Nathaniel Heintz, The Rockefeller University). After incubation with primary antibodies, the cells were incubated with secondary antibody conjugated to Alexa Fluor 488, Alexa Fluor 546, or Cy3 (anti-IgG, 1:1,000; Invitrogen). Cells were incubated with Hoechst 33258 (0.5 µg/ml in phosphate-buffered saline; Sigma-Aldrich) to stain nuclei. Cell counts were performed using a Nikon fluorescence microscope (x20 or x40 objective; Nikon, Tokyo, http://www.nikon.com) and Metamorph imaging software. Quantification of cells was based on counting the number of Hoechst-stained nuclei and the specific immunostained cells in independent fields from a minimum of three coverslips.
Cell Survival and Apoptotic Assays
A lactate dehydrogenase (LDH) Assay Kit was purchased from Promega (Madison, WI, http://www.promega.com). Cells were plated as a monolayer into a 24-well plate using the dissociation and plating methods listed above. Cells were plated in the presence of high or low EGF overnight, and then all medium was exchanged. Cells were maintained in the presence of high or low EGF for an additional 3 days prior to collection of media and lysing of the plated cells for maximal LDH release. Media and lysates were added to a 96-well microplate and were subjected to the manufacturer's protocol for assaying for LDH release. Optimal density of each sample was measured using a Molecular Devices SpectraMax 340pc system at 490 nm. The equation for calculating percentage of cytotoxicity is experimental LDH release divided by maximum LDH release. Verification of cell viability in both high- and low-EGF cultures was established by a trypan blue exclusion assay. Cells were plated using a method similar to those used for the LDH assay, only the cells were dissociated from the monolayer and subjected to trypan blue (0.4%; Sigma-Aldrich) staining. The numbers of positive and negative cells were established via counting using a light microscope. For apoptosis assay using neurosphere sections, neurospheres were allowed to settle in 15-ml Falcon tubes for 5 minutes before the culture medium was removed and replaced with 4% PFA for 20 minutes at room temperature. The fixed spheres were cryoprotected with 30% sucrose for 4 hours at 4°C. Finally, the spheres were embedded in Histoprep Frozen Tissue Embedding Media (Fisher Scientific, Fair Lawn, NJ, https://new.fishersci.com) compound, sectioned at 10 µm on a cryostat, and processed for cell death analysis using the DeadEnd Fluorometric terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) system (Promega).
Western Blotting and Phosphorylation Assays
Preparation of cell lysates and Western blotting were performed on neurospheres as described [6, 8]. Whole neurospheres were incubated in lysis buffer for 15 minutes on ice. For phosphorylation assays, neurospheres were dissociated and plated as a monolayer; they were subsequently subjected to growth factor treatment for 0, 5, 30, or 90 minutes or 24 hours and then lysed. Total protein concentrations were determined by using a range of bovine serum albumin standards in conjunction with a Bio-Rad protein analysis kit (Bio-Rad, Hercules, CA, http://www.bio-rad.com). Samples were read using a Molecular Devices SpectraMax 340pc system. Approximately 10 µg of protein was loaded into each well of a 10% gel for SDS-polyacrylamide gel electrophoresis. The gel was then transferred for 1 hour at 100 V to a polyvinylidene difluoride membrane (Bio-Rad). The membrane was then blocked with a 5% blotting solution and exposed to primary antibodies overnight at 4°C or 1 hour at room temperature. Secondary antibodies were applied for 1 hour at room temperature, and blots were exposed to enhanced chemiluminescent protocols (Amersham ECL Plus Western Blotting System, GE Healthcare, Buckinghamshire, UK, http://www.gehealthcare.com). Primary antibodies were as follows: BLBP, phosphorylated Akt (pAkt) (rabbit polyclonal; Upstate, Charlottesville, VA, http://www.upstate.com), epidermal growth factor receptor (EGFR; ErbB1), phosphorylated ErbB2 (pErbB2; rabbit polyclonal; Cell Signaling Technology, Danvers, MA, http://www.cellsignal.com), Actin (mouse monoclonal IgG2a; Sigma-Aldrich), S100 (rabbit polyclonal; Abcam, Cambridge, MA, http://www.abcam.com), Vimentin (mouse monoclonal IgM; Developmental Studies Hybridoma Bank, Iowa City, IA, http://www.uiowa.edu/
dshbwww), and GFAP (mouse monoclonal IgG1; Invitrogen, South San Francisco, CA). Blots were quantitated using the Scion Image software (Scion Corporation, Frederick, MD, http://www.scioncorp.com).
Reverse Transcription-Polymerase Chain Reaction
Total RNA was extracted from whole neurospheres grown in either high or low EGF using RNeasy purification systems (Qiagen, Valencia, CA, http://www1.qiagen.com). Reverse transcription-polymerase chain reaction (RT-PCR) was run for a maximum of 30 cycles on an MJ Research PT-200 thermal cycler (Bio-Rad). Primers were obtained from Integrated DNA Technologies (Coralville, IA, http://www.idtdna.com/Home/Home.aspx) and were combined with PCR Master Mix (2x; Promega). All primers were prepared from cDNA sequences for EGFR (ErbB1), ErbB2, ErbB3, ErbB4, Neuregulin-1 (NRG-1), Notch1, FGF-2, Glypican-1, Glypican-4, Pleiotrophin, fibroblast growth factor receptor 1 (FGFR1), and glyceraldehyde-3-phosphate dehydrogenase (as an internal control), which were obtained from the GenBank database (Table 1). The sequences were designed to cross an intron-exon boundary to prevent a false-positive signal due to genomic DNA.
|
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
To establish whether we were using optimal concentrations of EGF, we applied either 20 or 100 ng/ml to sister cultures at this stage. Volumetric measurements of single spheres over the course of 2 weeks revealed that those cultured in high EGF had a significantly greater volumetric increase than did neurospheres cultured in low EGF (Fig. 1A). Although 100 ng/ml EGF could increase the expansion rate of the cells, we still found that they entered a senescence phase between 50 and 70 weeks in vitro (data not shown), as described previously for these cells grown in 20 ng/ml EGF [7].
Given that high EGF promoted significant volumetric increases of hNPC-derived neurospheres, we next evaluated whether high EGF did this through increased proliferation or survival. Intact neurospheres were pulsed with BrdU for 14 hours while growing in the presence of EGF. These same neurospheres were subsequently dissociated and acutely plated for 2 hours in the absence of exogenous growth factors on laminin-coated coverslips. Immunostaining revealed no significant difference in BrdU incorporation between either treatment groups (Fig. 1B). Shorter, 3-hour pulses of BrdU (11.9% ± 1.4% of total cells in low EGF and 11.6% ± 1.3% in high EGF) or the use of other cell proliferation markers, such as Ki67 (data not shown), yielded similar results. Thus, high EGF did not promote increased neurosphere growth by increasing division rates. We therefore hypothesized that high the EGF was protecting progenitor cells after division.
We next asked whether high EGF was promoting increased survival of hNPC compared with cells treated with low EGF using an LDH assay. LDH is released from dying cells or cells with a compromised cell membrane and is commonly used as a marker of cell death. Because neurospheres are tightly interconnected three-dimensional structures that may not permit the release of LDH into the medium without other cells taking in the biochemical, we grew hNPC as a monolayer for 4 days in the presence of either high or low EGF. The culture medium was changed 1 day after dissociation/plating to allow the cells to recover from the acute neurosphere dissociation. We found a reduction of approximately 50% in the LDH released into the medium from cells treated with high EGF compared with the level in low EGF (Fig. 1C). Because LDH could have been released from cells with a compromised cell membrane but not necessarily from dying cells, we verified by a trypan blue exclusion assay that cells in the low-EGF treatment group had increased cell death rates. Overall, roughly 50% fewer cells from the high-EGF treatment group stained positive for trypan blue (Fig. 1D). Although the LDH and trypan blue exclusion assay were demonstrated with monolayered culture, we detected apoptotic cells in intact neurospheres to further assess the effects of high EGF. Apoptotic cells were detected by TUNEL assay using cryosections obtained from neurospheres grown for 14 days in either 20 or 100 ng EGF. High EGF treatment significantly reduced the number of TUNEL-positive cells, from 8.3% to 3.7% (P < 0.01; Fig. 1E). Taken together, these results supported our hypothesis that high EGF protects cells from imminent cell death following cell division.
We next aimed to establish a mechanism for increased hNPC survival in the presence of high EGF. EGF signals through the ErbB family of receptors. Autophosphorylation or upregulation of ErbB2 receptor activity has been associated with increased cell survival through phosphorylation of downstream elements such as Akt [20]. Our previous gene array studies suggested that ErbB1-ErbB3 receptors were expressed in hNPC, but ErbB4 was not [6]. It seemed reasonable to hypothesize that stimulation of hNPC with high EGF would increase expression of certain members of the ErbB family of receptors. Semiquantitative RT-PCR revealed that EGFR (ErB1) and ErB2 mRNA were expressed in all cultures and not differentially expressed in the two treatment groups (Fig. 1F). In addition, we confirmed that NRG-1 and Notch1 mRNA were expressed by hNPC (Fig. 1F) but not changed by high levels of EGF.
Only two members of the ErbB family are capable of binding EGF with any affinity: EGFR (ErbB1) and ErbB2 [21]. Interestingly, ErbB2 is not capable of binding EGF as a homodimer, so it must form heterodimers with EGFR. This alteration in ligand binding with two different ErbB subunits leads to differences in the intensity as well as the type of downstream signal elicited. Because ErbB2 is a potent signaler of the PI-3Kinase/Akt survival pathway, it seemed reasonable to hypothesize that high concentrations of EGF may stimulate a larger number of ErbB1/2 heterodimers, leading to a potent survival signal via Akt phosphorylation. Initially, evaluation of protein expression of pAkt and pErbB2 after a 5-minute pulse with EGF showed no significant difference between treatment groups (data not shown). This result was not surprising, as activation of downstream signals via the ErbB family requires times greater than 5 minutes [22]. Therefore, examination of a time course for EGF pulsing (0, 5, 30, and 90 minutes and 24 hours) was appropriate. A concentration of 100 ng/ml EGF, but not 20 ng/ml, elicited a prolonged, sustained pAkt signal, suggesting that stimulation via high EGF influences cell survival through this mechanism (Fig. 1G). This prolongation of Akt phosphorylation was accompanied by decreases in EGFR protein levels and increases in ErbB2 phosphorylation (Fig. 1G), supporting the idea that high EGF activates a different EGF signaling pathway. It was possible that this lack of sustained Akt phosphorylation at 20 ng/ml was simply due to the uptake of EGF by the dividing cells, thus exhausting the amounts available to the receptor. To answer this possibility, we cultured neurospheres in the medium with 20 ng/ml for 3 days (the same length of time at which we normally replenished the fresh medium) and then determined EGF levels in the culture medium by enzyme-linked immunosorbent assay. Compared with the EGF levels in preculture medium (17.7 ± 1.5 ng/ml), the EGF concentration after 3 days of culturing was 12.5 ± 0.2 ng/ml, suggesting that the concentration was still maintained at between 10 and 20 ng/ml.
Increased Elongated Cell Types Present in High-EGF hNPC Cultures
A small number of TuJ1-positive neuroblasts are observed within hNPC neurospheres, although the majority of cells express both nestin and GFAP [6]. By growing cells in a higher concentration of EGF, it seemed reasonable to hypothesize that more cells within the core of the neurosphere would be stimulated to divide as nestin/GFAP-positive progenitors and not divide to form neuroblasts [23, 24]. To investigate this possibility, hNPC were cultured in the medium with 20 or 100 ng/ml EGF for 14 days. The neurospheres were then dissociated to a single-cell suspension and plated for 1 hour before fixation and staining for the neuronal marker TuJ1, the astrocyte/progenitor cell marker GFAP, and the progenitor cell marker nestin. In keeping with our previous results [6, 8, 25], in 20 ng/ml EGF, the number of TuJ1-positive cells was less than 3% of the population, whereas GFAP- or Nestin-positive cells represented more than 75% of the population (Fig. 2A–2C). This was not changed by increasing the EGF concentration to 100 ng/ml (Fig. 2A–2C), showing that high EGF does not eliminate the small number of spontaneously differentiating neurons within the neurosphere.
|
To establish whether high EGF altered the presence of an intermediate neuronal progenitor cell population, we determined absolute protein levels of radial glial cell markers in response to EGF by using Western blot. Neurospheres were grown in the presence of high or low EGF for 14 days, followed by protein isolation. Western blot analysis revealed that high EGF increased the protein expression of gliogenic markers, such as radial glial markers BLBP and Vimentin and astroglial/radial glial marker S100 (Fig. 2H). Using the densitometric analysis of each band, we found that high EGF could increase BLBP, Vimentin, and S100 protein by 60%, 122%, and 23%, respectively, compared with low EGF. In contrast, the level of GFAP expression was not significantly different between high- and low-EGF cultures.
Growth of Neurospheres in High EGF Leads to Increased Neurogenesis Following Differentiation
Because radial glial cells are the primary neuron-producing cells in the developing cortex, we next established whether the observed increase in the radial glial-like morphology in the high-EGF culture led to an increase in neurogenesis upon differentiation. Cultures grown in 20 or 100 ng/ml EGF were dissociated, plated as a monolayer at high density (1,000 cells per microliter), and allowed to differentiate over 7 days as described in detail previously [6, 8, 25]. Cells derived from 100 ng/ml EGF cultures displayed increased neurogenesis compared with cells from 20 ng/ml EGF, whereas the number of GFAP-positive cells remained stable (Fig. 3A–3D). It remained possible that high EGF was not only stimulating hNPC to be maintained as radial glial cells but that it also converted some cells to intermediate neuronal progenitor cells.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
EGF signals through the ErbB family of receptors. Furthermore, ErbB1 and ErbB3 receptors have previously been shown to be responsible for activating radial glia and thus making them capable of producing neurons and allowing neuronal migration into the cortex along their long processes [28, 29]. Thus, it is possible that adding a high concentration of EGF to hNPC stimulated NRG-1 release and subsequent binding to ErbB2/ErbB3 heterodimers. However, because similar levels of ErbB3 and NRG-1 mRNA were observed in both high and low EGF, it is unlikely that NRG-1 binding to ErbB2/ErbB3 heterodimers was responsible for any of the observed effects in these cultures. The potential for ErbB2 activation, however, was also possible through ErbB1/ErbB2 heterodimerization. Sustained ErbB2 activation was found in hNPC treated with high EGF but not 20 ng/ml EGF. This survival signal was likely protecting new radial glial-like cells in the high-EGF-treated hNPC cultures, as we saw a reduction in cell death with an increase in the number of elongated cell types. This survival signal was likely a phosphorylated form of Akt, the antiapoptotic protein.
Although radial glial cells have long been known to produce cortical astrocytes in the developing brain, recent data indicate that radial glial cells might also possess neurogenic potential and divide asymmetrically to produce cortical neurons [11, 14]. In the presence of high levels of EGF, our cultures maintained a larger number of cells with a radial glial morphology. Although it has been difficult to find immunocytochemical markers that specifically identify these cells under the microscope, our Western analysis suggested that at least three markers of radial glia (BLBP, Vimentin, and S100) are increased in high EGF. When removed from the mitogen EGF, these radial glial-like cells undergo further division that produces neurons in this culture system [23, 24], thus leading to a higher proportion of neurons. Therefore, high EGF concentrations both increase the rate of division, allowing large numbers of cells to be generated, and enhance neuronal output from these cultures.
|
CONCLUSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H.-J. Kim, E. McMillan, F. Han, and C. N. Svendsen Regionally Specified Human Neural Progenitor Cells Derived from the Mesencephalon and Forebrain Undergo Increased Neurogenesis Following Overexpression of ASCL1 Stem Cells, February 1, 2009; 27(2): 390 - 398. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
