Another study showed a method to prevent tumor formation in monkeys:
http://stemcells.alphamedpress.org/c...005-0391v1.pdf
another article:
Journal of Neuroscience Research
RSS feed for Journal of Neuroscience Research What is RSS?
Volume 83, Issue 6, Pages 1015-1027
Published Online: 21 Feb 2006
Research Article
Generation of graftable dopaminergic neuron progenitors from mouse ES cells by a combination of coculture and neurosphere methods
Asuka Morizane 1, Jun Takahashi 1 *, Mizuya Shinoyama 1, Makoto Ideguchi 1, Yasushi Takagi 1, Hitoshi Fukuda 1, Masaomi Koyanagi 1, Yoshiki Sasai 2, Nobuo Hashimoto 1
Parkinson's disease is characterized by a loss of midbrain dopamine (DA) neurons and is generally viewed as a potential target for stem cell therapy. Although several studies have reported the generation of postmitotic DA neurons from embryonic stem (ES) cells, it is unknown whether the proliferative progenitors of DA neurons can be isolated in vitro. To investigate this possibility, we have developed a combined approach in which ES cells are cocultured with PA6 stromal cells to expose them to stromal cell-derived inducing activity (SDIA) and are then cultured as neurospheres.
Mouse ES cell colonies were detached from PA6 feeder cells after 8 days of SDIA treatment and then expanded as spheres for another 4 days in serum-free medium supplemented with fibroblast growth factor-2. The spheres exhibited neural stem cell characteristics and contained few DA neurons at this stage of culture. After being induced to differentiate on polyornithine/laminin-coated dishes for 7 days, these spheres generated DA neurons in vitro at a relatively low frequency.
Intriguingly, addition of PA6 cell conditioned medium to the sphere culture medium significantly increased the percentage of DA neurons to 25-30% of the total number of neurons. Transplantation of conditioned medium-treated day 4 spheres, which contained DA neuron progenitors, into the mouse striatum resulted in the generation of a significant number of graft-derived DA neurons. These findings suggest that progenitors of DA neurons are generated and can proliferate in ES cell-derived neurospheres induced by serial SDIA and PA6 conditioned medium treatment.
DISCUSSION
Here, we demonstrate that differentiating ES cells can be enriched for neural progenitor cells by culturing them on PA6 feeder cells, which provide a source of SDIA, and by using a sphere culture method. The addition of PA6 cell CM into the sphere culture medium enriches for proliferating DA neuron progenitors in spheres and results in the efficient differentiation of ES cells into DA neurons both in vitro and in vivo (Fig. 7). Thus, this method produces an efficient graftable cell source for cell transplantation. We also clearly demonstrate that the proliferating cells that gave rise to tumors were derived from undifferentiated ES cells or early ectodermal cells.
Figure 7. In vitro neural induction and dopaminergic differentiation by a combination of SDIA treatment and sphere culture. Neural induction was achieved by coculturing ES cells on PA6 feeder cells. By sphere culture in serum-free medium supplemented with B27 and FGF2, the cells proliferate while retaining characteristics of neural precursors. The CM of PA6 cells supports the survival or proliferation of DA neuron progenitors in spheres, resulting in efficient dopaminergic differentiation in vivo after transplantation.
[Normal View 37K | Magnified View 112K]
ES cell colonies growing on PA6 feeders are heterogeneous and consist of undifferentiated ES cells, neural stem/progenitor cells, differentiated neural cells, and nonneural cells, such as epidermal cells. When colonies are maintained on noncoated dishes in a serum-free medium supplemented with B27 and FGF2, neural stem/progenitor cells form spheres and proliferate in a manner comparable to that of mouse brain-derived neural stem cells (Gritti et al., [1996]), whereas other cell types tend to adhere to the bottom of the culture dish. Insofar as serum-free medium is not permissive for nonneural cell growth and ES cells are diverted into a neural lineage in the absence of LIF, it seemed likely that neural cell types would be enriched by the sphere culture method. Indeed, most of the cells in the spheres were immunoreactive for nestin or other neural cell markers and gave rise to neurons, astrocytes, and oligodendrocytes after differentiation. Previous work indicated that spheres exhibiting neural stem cell characteristics could be obtained from single mouse ES cells in the presence of leukemia inhibitory factor (LIF; Tropepe et al., [2001]) and from human embryoid bodies in the presence of FGF2 (Reubinoff et al., [2001]; Zhang et al., [2001]). In addition, dissociated mouse or human ES cells cultured on the bone marrow-derived stromal feeder cell line MS5 can form similar spheres in the presence of FGF2 (Barberi et al., [2003]; Perrier et al., [2004]).
When spheres were cultured at low density, the percentage of sphere-derived TH-positive cells decreased significantly. This finding suggests that ES cell-derived neuropheres may secrete soluble factors that promote dopaminergic differentiation, such as Wnts (Castelo-Branco et al., [2003]). Interestingly, among the soluble factors we tested, the CM of PA6 cells most efficiently induced DA neurons from spheres, yielding a percentage of TH-positive cells similar to that with continuous coculture on PA6 feeder cells. The CM increased the expression of En1 and En2, which are expressed in proliferating DA neuron progenitors (Arenas, [2002]; Burbach et al., [2003]), and mildly decreased expression of an earlier midbrain marker, Pax2, in CM spheres compared with control spheres. Thus, the CM of PA6 cells seems to promote dopaminergic differentiation or to support proliferation and/or survival of DA neuron progenitors. Although the molecular nature of SDIA is unknown, two possibilities have been proposed (Kawasaki et al., [2000]). One is that SDIA consists of two different neural-inducing factors, one that is anchored to the cell surface and one that is secreted. Another scenario is that secreted factors that are secondarily tethered to the cell surface are responsible for the activity. The facts that PA6 cells retained neural-inducing activity even after being fixed with paraformaldehyde and their CM could not elicit significant neural induction (Kawasaki et al., [2000]) suggest that there are two steps for the generation of DA neurons from ES cells: neural induction and maturation of DA neurons. The present results suggest that, although membrane-bound factor(s) might be necessary for neural induction, soluble factor(s) secreted from the ES cell-derived neurospheres and/or PA6 cells are sufficient for proliferation and/or survival of DA neuron progenitors.
FGF8 and Shh have been shown to be important for the production of DA neurons at the mid-/hindbrain boundary during development (Ye et al., [1998]). These factors also induce differentiation of DA neurons from mouse ES cells when added during the expansion phase of nestin-positive cells (Lee et al., [2000]) or when ES cells are grown on MS5 feeder cells (Barberi et al., [2003]). ES cell-derived neural progenitors express the FGF8 receptor FGFR3 as well as the Shh receptors Ptc and Smo (Lee et al., [2000]). The mild increase in the TH/TuJ1-positive cell ratio that was observed when the spheres were treated with FGF8 and Shh is compatible with these earlier findings. In addition, 17-estradiol has been reported to play a versatile role in the differentiation/maturation and survival of DA neurons (Behl, [2002]; Sawada et al., [2002]). Although the effect of estradiol on ES cells remains unknown, rat neural stem cells express estrogen receptors and (Brännvall et al., [2002]). IL-1 and GDNF were also reported to promote dopaminergic differentiation from rat mesencephalic progenitors (Ling et al., [1998]), and GDNF has been shown to be a survival factor for DA neurons (Lin et al., [1993]; Beck et al., [1995]; Tomac et al., [1995]). The spheres that we derived from SDIA-treated ES cells have characteristics of neural stem cells or DA neuron progenitors, so it is possible that 17-estradiol, IL-1, and GDNF play a role in the induction and maintenance of DA neurons.
As mentioned above, CM of PA6 cells was able to enrich for DA neuron progenitors in spheres, resulting in good survival of DA neurons in vivo. If the donor cells are too mature, as were the colonies used in the present study, the survival rate of the DA neurons decreases, likely because mature neurons are more vulnerable to mechanical damage, inflammatory cytokines, and neurotrophic factor insufficiency. Thus, the developmental stage of the donor cells is one of the keys to successful transplantation. Among the 20,000 cells injected into the mouse striatum, we observed 1,506 surviving TH-positive cells in the graft, or 75.3 TH-positive cells per 1,000 cells grafted. This figure is almost same as that from the previous report of mouse ES cell allografts (75.0 by Barberi et al., [2003]). These numbers are higher than those from allografts of rat embryonic ventral mesencephalon, for which efforts were made to improve cell survival using 21-aminosteroids (34.6 by Nakao et al., [1994]), GDNF infusion (29.9 by Rosenblad et al., [1996]), caspase inhibitor (24.0 by Schierle et al., [1999]), or simple allografts of rat neural stem cells (3.5 by Studer et al., [1998]; 0.8 by Carvey et al., [2001]). Furthermore, all the experiments except for ours used 6-OHDA-lesioned animals as hosts, in which depletion of DA neurons within the substantia nigra might promote DA neuronal differentiation and survival (Nishino et al., [2000]). As shown here, ES cells can be manipulated to contain a substantial number of proliferating progenitors, resulting in the generation of abundant TH-positive cells in vivo.
Our in vivo studies revealed that a decrease in the number of undifferentiated ES cells in the spheres was correlated with an absence of tumor formation. Previous efforts to transplant naïve mouse ES cells have resulted in the formation of teratomas or teratocarcinomas, even in the case of xenografts (Björklund et al., [2002]; Erdö et al., [2003]). In contrast, when mouse ES cells were differentiated before transplantation by stepwise treatment with cytokines (Kim et al., [2002]) or by culturing them on MS5 feeder cells in the presence of cytokines (Barberi et al., [2003]), the donor cells did not form any tumors. In the allograft of naïve mouse ES cells, transplants of 4,000 cells led to abundant tumor formation, whereas transplants of 400 cells showed a propensity to form teratomas only in the somatosensory cortex (Harkany et al., [2004]). These findings suggest that the cause of tumor formation is undifferentiated ES cells. We first demonstrated this clearly by showing the colocalization of SSEA-1 and E-cadherin in Ki67-positive proliferating cells. All the previous tumors derived from ES cells have been described as teratomas or teratocarcinomas (Björklund et al., [2002]; Erdö et al., [2003]; Harkany et al., [2004]; Fukuda et al., [2005]). The tumors in the present study, however, were undifferentiated or intermediately differentiated neuroectodermal tumors, probably because we grafted spheres of predifferentiated cells instead of using low- or single-cell-density grafts. We suggest that undifferentiated ES cells might be influenced by the surrounding neural cells in the sphere, giving rise to neural stem/progenitor cells in vivo. In this context, it is noteworthy that Musashi1-positive cells were observed around the Ki67/E-cadherin/SSEA-1-positive cells.
In summary, the combination of SDIA treatment and sphere culture with the CM of PA6 cells significantly increased the number of induced DA neurons in vitro and in vivo by enriching for their proliferating progenitors in the spheres. However, it also partially restored the propensity of these cells to form tumors. Single ES cells are able to proliferate on PA6 feeders, so it is possible that the PA6 cell CM supported the survival and/or proliferation of not only DA neuron progenitors but also undifferentiated ES cells. As a result of these findings, the determination of the molecular mechanism of SDIA and the complete elimination of undifferentiated ES cells at an early stage of neural induction (Fukuda et al., [2005]) will be indispensable for the successful transplantation of ES cell-derived neural stem/progenitor cells.
full article:
http://www3.interscience.wiley.com/c...8892/HTMLSTART