Successful elimination of non-neural cells and unachievable elimination of glial cells by means of commonly used cell culture manipulations during differentiation of GFAP and SOX2 positive neural progenitors (NHA) to neuronal cells
© Witusik et al; licensee BioMed Central Ltd. 2008
Received: 25 March 2008
Accepted: 19 July 2008
Published: 19 July 2008
Although extensive research has been performed to control differentiation of neural stem cells – still, the response of those cells to diverse cell culture conditions often appears to be random and difficult to predict. To this end, we strived to obtain stabilized protocol of NHA cells differentiation – allowing for an increase in percentage yield of neuronal cells.
Uncommitted GFAP and SOX2 positive neural progenitors – so-called, Normal Human Astrocytes (NHA) were differentiated in different environmental conditions to: only neural cells consisted of neuronal [MAP2+, GFAP-] and glial [GFAP+, MAP2-] population, non-neural cells [CD44+, VIMENTIN+, FIBRONECTIN+, MAP2-, GFAP-, S100β-, SOX2-], or mixture of neural and non-neural cells.
In spite of successfully increasing the percentage yield of glial and neuronal vs. non-neural cells by means of environmental changes, we were not able to increase significantly the percentage of neuronal (GABA-ergic and catecholaminergic) over glial cells under several different cell culture testing conditions. Supplementing serum-free medium with several growth factors (SHH, bFGF, GDNF) did not radically change the ratio between neuronal and glial cells – i.e., 1,1:1 in medium without growth factors and 1,4:1 in medium with GDNF, respectively.
We suggest that biotechnologists attempting to enrich in vitro neural cell cultures in one type of cells – such as that required for transplantology purposes, should consider the strong limiting influence of intrinsic factors upon extracellular factors commonly tested in cell culture conditions.
So-called NHA (Normal Human Astrocytes) cells belong to the class of GFAP-positive neural progenitors. Expression of neuronal and glial markers during differentiation of these cells is regulated in accordance with the "model of discordant phenotypes suppression" [1–3]. This model states that before differentiation, markers belonging to diverse lineages are expressed by stem cells or progenitors; whereas during differentiation, genes superfluous in derived lines are silenced. In accordance with this concept, uncommitted neural progenitors (NHA) co-express glial [GFAP, CD44], neuronal [β-III-TUBULIN, MAP2] and progenitor [NESTIN] markers. GFAP positive neural progenitors (NHA) derivatives differentiating to the neuronal lineage demonstrated silencing of glial and progenitor markers [GFAP, CD44, NESTIN] expression, while those differentiating to the glial lineage showed silencing of neuronal and progenitor markers [β-III-TUBULIN, MAP2, NESTIN] expression. However expression of MAP2 and β-III-TUBULIN is definitely increased in neuronal derivatives of NHA [1, 2].
We could not prove so far that the coexpression of glial and neuronal markers is a consequence of physiological process. To this end we consider our investigation as biotechnologically useful, and we do not imply that a similar scenario has to occur in vivo. However report revealing presence of neural stem cells or progenitors expressing neuronal markers in vivo has been published lately by Walker et al . Coexpression of glial and neuronal markers in neural progenitors was also presented . Moreover, article showing human fetal astrocytes coexpressing in vivo GFAP, β-III-TUBULIN and MAP2 was published . The expression of β-III-TUBULIN in conjunction with MAP2 in GFAP-positive radial glia has been considered as suggestion of neuronal-glial bipotentiality . Radial glia are recognized as cells presenting NSCs properties [7, 8].
According to the suppression of discordant phenotypes model, the expression of markers characteristic for particular lineages in progenitor cells allows for the presumption of potential derivatives that can be obtained after differentiation. This model, however, is of very limited help in the biotechnological regulation of differentiation. Hence, we took into account others such as the instructive, stochastic, and continuum models – considering the possibility that they can help to increase the percentage yield of a given required cell type via cell culture manipulations.
The instructive (deterministic) model recognizes growth factors as elements which determine the fate of stem cells – thereby, triggering a particular differentiation pathway [9, 10]. Extensive research has shown that the fate of stem cells can be influenced by exogenous factors. However, their response to environmental signals often emerge to be random and difficult to predict. To this end, biologists have realised that an alternative model to the deterministic one would be required to explain better how the differentiation process is regulated. Hence, difficulties in predicting the response of stem cells have inspired a consideration of the stochastic models .
A stochastic model was popularized amongst haematologists by papers such as the one by Enver titled "Do stem cells play dice?" . According to this view, it is stochastic events that trigger diverse intracellular programs to regulate the differentiation of cells. The stochastic model recognizes growth factors as the important but permissive regulators of differentiation that support the survival and proliferation of one or a few already determined cell type(s) . Neurobiologists had started acknowledging a long time ago that stochastic events might play an important role during differentiation . In time, however, deterministic models became more popular amongst them [15–17].
Another popular model of differentiation is the continuum model. This model refers to the phenotypical changes that occur in stem cells during the cell cycle . The continuum model is based on the assumption that periodical and fluctuating phenotypical changes in stem cells cause the inevitable heterogeneity of their population, which – in turn, is responsible for the difference in their response to the environment.
However, recognizing the very different nature of events that determines the fate of progenitors or stem cells, both – stochastic and continuum, models have identified intracellular events as having the limiting influence on extracellular factors.
The literature data showing that developmental options of very restricted precursors (lymphoblasts, glial restricted precursors), can be analysed as inclined (predisposed) by both extracellular and intracellular events [19, 20], empowered us to transfer similar experimental attempt to investigate differentiation of neural progenitor cells (NHA).
We report here that the simultaneous consideration of more than one differentiation model may help in (a) controlling (understanding) human neural progenitor cells (NHA) differentiation, and (b) recognizing the limitations of typical biotechnological manipulations of cell cultures.
Expansion of GFAP-positive neural progenitors (NHA)
NHA isolated from the cerebrum of human foetuses were purchased from Cambrex (CC-2565, NHA-Normal Human Astrocytes; Walkersville, MD). NHA cells were cultured by plating at a density of 10 000–15 000 cells/cm2 in expansion medium (Cambrex AGM Bullet Kit: 3% FBS – fetal bovine serum, rhEGF – recombinant human epidermal growth factor, insulin and ascorbic acid), in accordance with the manufacturer's protocol. The cells were incubated at 37°C with 5% humidified CO2 and allowed to reach confluence during 1–2 days. Once they have reached about 95% or 50% of confluence, the cells were then either used for differentiation purposes or were harvested with 0.25% trypsin and 1 mM EDTA for 5 min at 37°C counted, and passaged.
Protocols of differentiation
1. Differentiation of GFAP+ neural progenitors to non-neural cells
GFAP-positive neural progenitors (NHA) were plated on six well dishes or uncoated chambers (Nunc) at a density of 10 000–15 000 cells/cm2 and passaged in expansion medium containing 3% FBS, rhEGF, insulin and ascorbic acid (Cambrex AGM Bullet Kit). After every passage, the cells were cultured for 5 days before the next passage was performed.
2. Differentiation of GFAP+ neural progenitors to neuronal and glial populations
GFAP-positive neural progenitors (NHA) at the passage 0 were plated at a density of 10 000–15 000 cells/cm2 on Matrigel thin-coated (according manufacturer's protocol, BD Biosciences), chamber slides and cultured in expansion medium (Cambrex AGM Bullet Kit) until reaching confluence. After reaching confluence, the medium was changed to N2- or B27-supplemented DMEM F12 (Invitrogen). A final concentration of 10 ng/ml of growth factor was achieved with the optional addition of either bFGF – basic fibroblast growth factor, SHH – sonic hedgehog or GDNF – glial derived neutrophic factor (Invitrogen). The cells were then cultured under differentiation conditions for 5–7 days, with medium replacement every 2 days.
3. Differentiation of GFAP+ progenitors to neuronal, glial and non-neural cells
GFAP+ neural progenitors were plated on six well dishes or uncoated chambers at a density of 10 000–15 000 cells/cm2 and kept for one to two passages in expansion medium (Cambrex AGM Bullet Kit). After every passage, the cells were kept for 5 days before the next passage was performed. After reaching confluence, the medium was changed to N2- or B27-supplemented DMEM F12 (Invitrogen). A final concentration of 10 ng/ml of growth factor was achieved with the optional addition of one of the following growth factors: bFGF, SHH or GDNF (Invitrogen). The cells were then cultured under differentiation conditions for 5–7 days, with medium replacement every 2 days.
Low density seeding
The differentiation protocol 2 and protocol 3 were performed also after seeding of NHA cells passage 0 at the following densities: 7 000, 5 000, 2 500 and 1 000 cells/cm2.
After seeding, the cells were cultured in expansion medium (Cambrex AGM Bullet Kit) for 5 days, then the medium was changed to N2- or B27-supplemented DMEM F12 with bFGF (Invitrogen) for 5–7 days; the medium was replaced every 2 days.
Protocol testing commitment of NHA cells passage 0
Low density seeding used for subcloning purposes was established at 5 000 cells/cm2.
NHA cells were plated on Matrigel thin-coated (manufacturer's protocol; BD Biosciences) chamber slides and cultured in expansion medium (Cambrex AGM Bullet Kit) until they reached 50% or 95% confluence. Then the cells were grown for 5–7 days in N2- or B27-supplemented DMEM F12 with bFGF (Invitrogen), with medium replacement every 2 days.
Primary antibodies applied for immunocytochemistry purposes
Santa Cruz Biotechnology, Inc.; sc-6171
Santa Cruz Biotechnology, Inc.; sc-7297
Santa Cruz Biotechnology, Inc.; sc-20172
Sigma; F 3648
Santa Cruz Biotechnology, Inc.; sc- 25269
Pel-Freez Biologicals; P40101-0
Santa Cruz Biotechnology, Inc.; sc-23927
Santa Cruz Biotechnology, Inc.; sc-81709
Immunoassayed cells were analyzed with an Olympus BX-41 fluorescence microscope. Ten fields were counted for each cover-slip – with each field containing minimum 50 cells. The number of positive cells was averaged from at least three cover-slips from three independent experiments. The percentages of cells positive for MAP2, GFAP and/or CD44 were calculated after differentiation.
Real time PCR
The RNA samples were obtained from undifferentiated NHA cells and from non-neural population (differentiation protocol 1), neural cells (differentiation protocol 2) and population consists of both cell types (differentiation protocol 3).
Total cellular RNA was isolated from cell cultures using NucleoSpin RNA/Protein kit (MACHERY-NAGEL) according to the manufacturer's protocol. Contaminating residual genomic DNA was removed by DNase I digestion and an aliquot containing 1 μg of isolated RNA was reverse transcribed into single-stranded cDNA in a final volume of 40 μl using QuantiTect Reverse Transcription Kit (Qiagen).
A Real time quantitative PCR was performed on a Rotor Gene 6000 (Corbett, Life Sciences, Australia) instrument utilizing the SYBR® Green Master Mix (Applied Biosystems, CA). Following MAP2 primers were used for amplification:
F 5' GCACTTCAAGGGAAGCTGAT 3', R 5' ATCAAATGGTCCACTAGGCG 3'.
Each sample was amplified in triplicate (for each run) in a reaction volume of 10 μl containing 50 ng cDNA, 1× SYBR Green Master Mix and forward and reverse primer 35 ng each. The cycling conditions were as follows: 3 min at 95°C (polymerase activation) followed by 40 cycles of 15 s at 95°C (denaturation), 45 s at 59°C (annealing) and 1 min at 72°C (extension). ACTB was used as a reference gene for normalization of target genes expression levels. Normalized relative expression level for a given gene of non-neural cells, cell population consisted of neural and non-neural cells and neural population obtained after differentiation versus undifferentiated cells was calculated utilizing the method described previously by Pfaffl et al. and implemented into computer application REST2005 , based on each sample's average CT value and each gene's average PCR efficiency. Additionally, the Real time PCR products obtained after 20 cycles were analyzed on 2% agarose gel electrophoresis.
Western blot analysis
The protein samples were isolated from undifferentiated NHA cells and from non-neural population (differentiation protocol 1), neural cells (differentiation protocol 2) and population consists of both cell types (differentiation protocol 3).
The protein extracts from cell cultures were obtained with the use of NucleoSpin RNA/Protein kit (MACHERY-NAGEL). The commercially available Brain Tissue Lysate was purchased from Abcam and used as positive control in Western blot analysis. Equal amounts of protein extracts, 25 μg per line, were separated by 5% SDS-PAGE followed by transfer onto Immobilon™-P polyvinylidene difluoride membranes (Sigma-Aldrich). After blocking nonspecific binding the membranes were incubated for 1 hour at room temperature with rabbit anti-MAP2 primary antibody (Santa Cruz Biotechnology, Inc.; sc-20172). After extensive washing in TBST buffer (TRIS Buffered Saline with Tween 20; pH 7.6) the membranes were incubated for 1 hour at room temperature with anti-rabbit secondary antibody (Santa Cruz Biotechnology, Inc.), diluted 1:2000. Membrane was then washed in TBST buffer and finally in TBS buffer (TRIS Buffered Saline; pH 7.5); antigen-antibody complexes were detected by enhanced chemiluminescence, using Chemiluminescence Luminol Reagent (Santa Cruz Biotechnology, Inc.) and visualized with the use of BIO-RAD camera and Quantity One software.
The effect of cell density on non-neural population percentage was assessed by the non-linear estimation by fitting the experimental data to exponential curve. Log-linear analysis was employed in order to analyse effects of medium and growth factors used in cell culture to increase percentage of neuronal cells. The differences in gene expression levels were determined by the bootstrap randomization technique, implemented in program REST2005. Statistical significance was assumed for P-value ≤ 0.05.
NHA cells at the passage 0 showed one homogenous phenotype
Environmental factors affect the fate of NHA cells
Two completely different derivatives could be obtained following exposure of NHA cells to two diverse environments.
The proportion of neuronal and glial cells obtained in different culture conditions
NHA-passage 0; 5–7 days
DMEM F12 with N2
DMEM F12 with B27
MAP2+ GFAP- CD44-
GFAP+, CD44+, MAP2-
MAP2+, GFAP+, CD44-
NHA cells that were seeded at a density of 10 000–15 000/cm2, and exposed for one passage to expansion medium (Cambrex medium with 3% FBS, rhEGF, insulin, and ascorbic acid), until they reached confluence, and subsequently exposed to N2- or B27-supplemented DMEM F12, produced three main populations of cells: (1) glial [GFAP+, CD44+, MAP2-], (2) neuronal [MAP2+, CD44-, GFAP-], and (3) non-neural cells robustly producing FIBRONECTIN [CD44+, VIMENTIN+, GFAP-, MAP2-, S100β-, SOX2-] in brief [CD44+, GFAP-, MAP2-]. Only 5–10% of the cells presented phenotype [CD44+, GFAP-, MAP2-] after differentiation at the passage 1 (Figures 3 and 4; Protocol 3). NHA cells at the passage 3 produced approximately 40% of cells with [CD44+, GFAP-, MAP2-] phenotype and equal proportions of neuronal and glial cells (Figures 3 and 4; Protocol 3). Cells with phenotype [GFAP+, MAP2+ CD44-] were – as before (Table 2), rarely observed after following protocol 3.
Influence of seeding density on percentage of non-neural cells obtained after differentiation
Non-neural cells percentage
15 000 cells/cm2
10 000 cells/cm2
7 000 cells/cm2
5 000 cells/cm2
2 500 cells/cm2
1 000 cells/cm2
Changes in density seeding of NHA cells at the passage 0 also influenced the fate of GFAP-positive progenitors presented here (see below).
Decreasing seeding density of NHA cells at the passage 0 caused appearance of non-neural cells
NHA cells at the passage 0 were seeded at densities of 15 000 cells/cm2, 10 000 cells/cm2, 7 000 cells/cm2, 5 000 cells/cm2, 2 500 cells/cm2 and 1 000 cells/cm2. In spite of the fact that they were all at the passage 0, the cells that were seeded at densities lower than 7 000 cells/cm2 were observed to present [CD44+, GFAP-, MAP2-, SOX2-] phenotype, once confluence was reached in the Cambrex medium with 3% FBS, rhEGF and ascorbic acid, followed by seven days exposure to DMEMF12 with N2 (B27). Influence of seeding density on NHA cells passage 0 differentiation is presented in Table 3.
In order to evaluate the effect of cell density on percentage of non-neural population, experimental data were fitted to exponential curve y = 122.81 * exp(-0.0004x) (p < 0.0001). Regression analysis explained 90% of variability in percentage of non-neural population (r = -0.09509; R^2 = 0.9043). These results point to the significant influence of the environmental factor (seeding density) on NHA cells differentiation.
Subcloning of NHA cells do not suggest commitment before serum starvation
When colonies of neuronal and glial cells were observed, however, radical change in the differentiation protocol was required to obtain them. Medium with serum had to be replaced with DMEM F12 with N2 (B27) before reaching confluence by the original NHA cell culture (passage 0, seeding density 5 000 cells/cm2). Presence of colonies after that modification was a consequence of the fact that committed in DMEM F12 with N2 (B27) at low density stage glial and neuronal cells were proliferating within free of cells spaces forming colonies (Figure 6c). It proved that committed, proliferating cells forming colonies could be detected by means of presented here subcloning method.
GDNF, SHH and bFGF did not radically change the ratio between neuronal and glial cells
NHA cultured in DMEM F12 with N2 or B27 generated neuronal and glial cells. After 5–7 days of differentiation, we observed cell populations of 50% neuronal and 46% glial and approximately 4% of undifferentiated cells [GFAP+, MAP2+, CD44-] (Table 2). However, with the addition of either bFGF, GDNF, or SHH, the populations turned out to be 54–57% neuronal and 41–44% glial (Table 2).
Statistical analysis indicated that influence of growth factors and supplements used in cell culture to increase percentage of neuronal cells was not significant; (P = 1.00).
Presence of GABA and TH
The main goal of our current research was to obtain the highest possible percentage of biotechnologically useful-cetacholaminergic and GABA-ergic, neuronal cells via differentiation of NHA neural progenitors. To this end, we considered the influence of intrinsic and extrinsic factors during NHA cells differentiation. NHA cells have been described as GFAP-positive neural progenitors in our previous reports [1, 2], but here, for the first time, we showed that these cells express SOX2 at the passage 0. In spite of the fact, that expression of SOX genes are associated with neural stem cells and SOX proteins are regarded as markers of neural primordium , NHA differentiation status is not obvious. According to our results these cells were identified as neural progenitors, but we couldn't exclude a possibility that NHA may reveal features of stem cells [1, 2]. Anyway, the influence of intrinsic and extrinsic events should be considered not only in the context of stem cells, since also progenitors are biotechnologically valuable.
We found nothing to suggest that NHA cells culture at the passage 0 consists of two or more cell populations presenting permanent, diverse intrinsic regulators of differentiation. In other words, we found no data to support the idea that the fate of NHA cells is determined at the passage 0 before serum starvation (Figures 1 and 6) [1, 2]. However, we cannot exclude the possibility that NHA cells were changing their phenotype periodically and discretely in accordance with the continuum model . A hypothesis of the permanent fate pre-determination was also undermined by the results showing that uncommitted NHA cells were very selectively differentiated in serum-free conditions into neural cells. On the other hand, in medium containing serum, NHA cells were directed, after 4–5 passages or extremely low density seeding at the passage 0, with 100% efficiency towards cells recognized temporarily as non-neural on the basis of presented phenotype [CD44+, VIMENTIN+, FIBRONECTIN+, SOX2-, MAP2-, GFAP-, S100β-, NESTIN-, β-III-TUBULIN-] and disability to differentiate into the neuronal derivatives (Figure 5; for NESTIN and β-III-TUBULIN data not shown). Provided evidences entitled to regard this population as non-neural, nevertheless, defining exact phenotype of these cells requires additional investigation.
Our results showed that environmental changes triggered two different types of differentiation: neural and non-neural (Figures 3, 4 and 5). These data fit the instructive model of differentiation. It also showed, that time of exposure to a particular environment is important during the differentiation of NHA progenitors. The role of the duration of exposure to a particular environment in regulating differentiation has been recognized as a kinetic factor .
In spite of findings on the determination of neural versus non-neural fate by means of simple environmental manipulation, we were unable to obtain strictly neuronal or strictly glial cells from any of the cell culture conditions tested. This shows the inadequacy of the instructive model in explaining how glial vs. neuronal fate determination is regulated during GFAP positive neural progenitors (NHA) differentiation. We should ask "why" initially uncommitted NHA cells always end up differentiating into two types of neural cells – i.e., neuronal and glial, even when exposed to different – serum free environments. Stochastic and continuum models of differentiation do offer the answer to this question. In accordance with the stochastic model, stochastic events can trigger intracellular developmental diversity and cause the appearance of two (or more) different responses to one environment . Referring to the Enver suggestion that "hematopoietic stem cells play dice" , we may suggest that after serum starvation "NHA neural progenitors toss a coin and – getting a head, choose a neuronal fate; whereas getting tail meant choosing a glial fate." The continuum model – with its fluctuating reversible changes during cell cycle temporarily biasing the fate of progenitors, also fits the results that we observed. The stochastic or continuum nature of the process regulating neuronal vs. glial cells determination was supported by the results of our sub-cloning experiment (Figure 6). In essence, these models can explain our data, which shows that growth factors such as bFGF, GDNF, or SHH were not able to increase dramatically the percentage of neuronal cells within, five to seven days exposure to the serum free medium (Table 2). Both models are based on the thesis that intracellular events cause certain cellular independence in terms of their responses to environmental changes. In our experiments, growth factors were unable to trigger only neuronal differentiation from neural progenitors (NHA), which is in accordance with stochastic and continuum models.
It seems important, from the biotechnological point of view, that we were able to limit the percentage of non-neural cells [CD44+, VIMENTIN+, FIBRONECTIN+, GFAP-, MAP2-, S100β-, SOX2-] (Figures 3 and 4) since cells robustly producing FIBRONECTIN could be hazardous as an element of graft in the central nervous system [24, 25]. Moreover these cells were unable to provide neuronal cells and therefore we recognized them as useless considering our project target – the procurement of neuronal cells. In spite of the successful elimination of non-neural cells, we did not manage to increase significantly the ratio between neuronal versus glial cells (Table 2). It could be disappointing that we were almost unable to break the symmetrical proportion between neuronal and glial cells while testing several growth factors. Nevertheless, we can still propose an answer for the question of why either stochastic or continuum models, basing on the primary role of intrinsic factors and the permissive role of extrinsic factors, can describe the biologically designed process of neural progenitors (NHA) differentiation. It is very likely that in vitro glial cells – as in in vivo ones, support neuronal cell growth, differentiation, and maturation [26, 27]. Since the biological process regulating the differentiation of neural progenitors seems to be rational in its simultaneous provision of glial and neuronal cells, biotechnological manipulations leading to the obtainment of only – or almost only, neuronal cells could result in the elimination of natural supporters of neuronal cells.
The present study was supported by the Ministry of Scientific Research and Information Technology, Poland, grant No. N 401 046 31/1195; this founding covered all Materials and Methods payments.
Dr Ewa Golanska and Dr Giac Nguyen from the Medical University of Lodz are kindly acknowledged for language assistance.
- Rieske P, Azizi SA, Augelli B, Gaughan J, Krynska BA: Population of human brain parenchymal cells express markers of glial, neuronal and early neural cells and differentiate into cells of neuronal and glial lineages. Eur J Neurosci. 2007, 25: 31-7. 10.1111/j.1460-9568.2006.05254.x.View ArticleGoogle Scholar
- Witusik M, Gresner SM, Hulas-Bigoszewska K, Krynska B, Azizi SA, Liberski PP, Brown P, Rieske P: Neuronal and astrocytic cells, obtained after differentiation of human neural GFAP-positive progenitors, present heterogeneous expression of PrP(c). Brain Res. 2007, 1186: 65-73. 10.1016/j.brainres.2007.10.039.View ArticleGoogle Scholar
- Egusa H, Schweizer FE, Wang CC, Matsuka Y, Nishimura I: Neuronal differentiation of bone marrow-derived stromal stem cells involves suppression of discordant phenotypes through gene silencing. J Biol Chem. 2005, 280: 23691-7. 10.1074/jbc.M413796200.View ArticleGoogle Scholar
- Walker TL, Yasuda T, Adams DJ, Bartlett PF: The doublecortin-expressing population in the developing and adult brain contains multipotential precursors in addition to neuronal-lineage cells. J Neurosci. 2007, 27: 3734-42. 10.1523/JNEUROSCI.5060-06.2007.View ArticleGoogle Scholar
- Bantubungi K, Blum D, Cuvelier L, Wislet-Gendebien S, Rogister B, Brouillet E, Schiffmann SN: Stem cell factor and mesenchymal and neural stem cell transplantation in a rat model of Huntington's disease. Mol Cell Neurosci. 2008, 37: 454-70. 10.1016/j.mcn.2007.11.001.View ArticleGoogle Scholar
- Dráberová E, Del Valle L, Gordon J, Marková V, Smejkalová B, Bertrand L, de Chadarévian JP, Agamanolis DP, Legido A, Khalili K, Dráber P, Katsetos CD: Class III beta-tubulin is constitutively coexpressed with glial fibrillary acidic protein and nestin in midgestational human fetal astrocytes: implications for phenotypic identity. J Neuropathol Exp Neurol. 2008, 67: 341-54.View ArticleGoogle Scholar
- Zecevic N: Specific characteristic of radial glia in the human fetal telencephalon. Glia. 2004, 48: 27-35. 10.1002/glia.20044.View ArticleGoogle Scholar
- Mo Z, Moore AR, Filipovic R, Ogawa Y, Kazuhiro I, Antic SD, Zecevic N: Human cortical neurons originate from radial glia and neuron-restricted progenitors. J Neurosci. 2007, 27: 4132-45. 10.1523/JNEUROSCI.0111-07.2007.View ArticleGoogle Scholar
- Muller-Sieburg CE, Cho RH, Thoman M, Adkins B, Sieburg HB: Deterministic regulation of hematopoietic stem cell self-renewal and differentiation. Blood. 2002, 100: 1302-9.Google Scholar
- Kubota H, Avarbock MR, Brinster RL: Culture conditions and single growth factors affect fate determination of mouse spermatogonial stem cells. Biol Reprod. 2004, 71: 722-31. 10.1095/biolreprod.104.029207.View ArticleGoogle Scholar
- Abkowitz JL, Catlin SN, Guttorp P: Evidence that hematopoiesis may be a stochastic process in vivo. Nat Med. 1996, 2: 190-7. 10.1038/nm0296-190.View ArticleGoogle Scholar
- Enver T, Heyworth CM, Dexter TM: Do stem cells play dice?. Blood. 1998, 92: 348-51.Google Scholar
- Thornley I, Sutherland R, Wynn R, Nayar R, Sung L, Corpus G, Kiss T, Lipton J, Doyle J, Saunders F, Kamel-Reid S, Freedman M, Messner H: Early hematopoietic reconstitution after clinical stem cell transplantation: evidence for stochastic stem cell behavior and limited acceleration in telomere loss. Blood. 2002, 99: 2387-96. 10.1182/blood.V99.7.2387.View ArticleGoogle Scholar
- Baroffio A, Blot M: Statistical evidence for a random commitment of pluripotent cephalic neural crest cells. J Cell Sci. 1992, 103: 581-7.Google Scholar
- Vicario-Abejon C, Johe KK, Hazel TG, Collazo D, McKay RD: Functions of basic fibroblast growth factor and neurotrophins in the differentiation of hippocampal neurons. Neuron. 1995, 15: 105-14. 10.1016/0896-6273(95)90068-3.View ArticleGoogle Scholar
- Buytaert-Hoefen KA, Alvarez E, Freed CR: Generation of tyrosine hydroxylase positive neurons from human embryonic stem cells after coculture with cellular substrates and exposure to GDNF. Stem Cells. 2004, 22: 669-74. 10.1634/stemcells.22-5-669.View ArticleGoogle Scholar
- Fu M, Lui VC, Sham MH, Pachnis V, Tam PK: Sonic hedgehog regulates the proliferation, differentiation, and migration of enteric neural crest cells in gut. J Cell Biol. 2004, 166: 673-84. 10.1083/jcb.200401077.View ArticleGoogle Scholar
- Quesenberry PJ, Colvin G, Dooner G, Dooner M, Aliotta JM, Johnson K: The stem cell continuum: cell cycle, injury, and phenotype lability. Ann N Y Acad Sci. 2007, 1106: 20-9. 10.1196/annals.1392.016.View ArticleGoogle Scholar
- Nutt SL, Vambrie S, Steinlein P, Kozmik Z, Rolink A, Weith A, Busslinger M: Independent regulation of the two Pax5 alleles during B-cell development. Nat Genet. 1999, 21: 390-395. 10.1038/7720.View ArticleGoogle Scholar
- Busslinger M, Nutt SL, Rolink AG: Lineage commitment in lymphopoiesis. Curr Opin Immunol. 2000, 12: 151-8. 10.1016/S0952-7915(99)00065-5.View ArticleGoogle Scholar
- Pfaffl MW, Horgan GW, Dempfle L: Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res. 2002, 30: e36-10.1093/nar/30.9.e36.View ArticleGoogle Scholar
- Ellis P, Fagan BM, Magness ST, Hutton S, Taranova O, Hayashi S, McMahon A, Rao M, Pevny L: SOX2, a persistent marker for multipotential neural stem cells derived from embryonic stem cells, the embryo or the adult. Dev Neurosci. 2004, 26: 148-165. 10.1159/000082134.View ArticleGoogle Scholar
- Abkowitz JL, Golinelli D, Harrison DE, Guttorp P: In vivo kinetics of murine hemopoietic stem cells. Blood. 2000, 96: 3399-405.Google Scholar
- Kawano H, Li HP, Sango K, Kawamura K, Raisman G: Inhibition of collagen synthesis overrides the age-related failure of regeneration of nigrostriatal dopaminergic axons. J Neurosci Res. 2005, 80: 191-202. 10.1002/jnr.20441.View ArticleGoogle Scholar
- Stichel CC, Hermanns S, Luhmann HJ, Lausberg F, Niermann H, D'Urso D, Servos G, Hartwig HG, Müller HW: Inhibition of collagen IV deposition promotes regeneration of injured CNS axons. Eur J Neurosci. 1999, 11: 632-46. 10.1046/j.1460-9568.1999.00466.x.View ArticleGoogle Scholar
- Bouzier-Sore AK, Merle M, Magistretti PJ, Pellerin L: Feeding active neurons: (re)emergence of a nursing role for astrocytes. J Physiol Paris. 2002, 96: 273-82. 10.1016/S0928-4257(02)00016-5.View ArticleGoogle Scholar
- Ma DK, Ming GL, Song H: Glial influences on neural stem cell development: cellular niches for adult neurogenesis. Curr Opin Neurobiol. 2005, 15: 514-20. 10.1016/j.conb.2005.08.003.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.