The Niche

Research round-up: self-oranizing ES cells; more reprogrammig molecules

Lots of papers out recently. These will appear as separate articles on Thursday on the home page. Here are the titles ow. The articles are below.

Human cortical neural cell balls: Cells cultured in embryoid bodies self-organize into specific neural subtypes

Exogenous aids to reprogramming and self-renewal: Small molecules replace pluripotency gene and Wnt boosts self-renewal and reprogramming

Embryoid bodies get organized: Wnt signalling stimulates the beginnings of gastrulation

Human cortical neural cell balls: Cells cultured in embryoid bodies self-organize into specific neural subtypes

Embryonic stem cells can be readily coaxed into becoming neural progenitors, but neurons from the cerebral cortex have proved extremely difficult to make in vitro, presumably because forming these neurons requires input from other cells in a developing brain. Earlier this year, a group led by Pierre Vanderhaeghen of the Institute for Interdisciplinary Research at the University of Brussels, Belgium, showed how to make these cells in a flat monolayer culture1. Now, a team lead by Yoshiko Sasai at the RIKEN Center for Developmental Biology in Kobe, Japan, shows that both mouse and human embryonic stem cells can organize themselves in culture into three-dimensional structures that recapitulate processes in early development to form cortical neural cells.2

“Both are fantastic papers,” says Stewart Anderson, a developmental neurobiologist turned stem cell scientist at Weill Cornell Medical College in New York, but he added that the culture system from the Sasai team promises more applications. “They can get these growing balls of cortical cells such that they can try to make different regions of cells depending on extracellular signals,” he says. “If the Vanderhaeghen group has made a really nice BMW, the Sasai group has made a Ferrari.”

Anderson cautions that these culture-derived cells still need to be more thoroughly compared to their endogenous counterparts. Sasai identified cell types with markers and used transplantation studies to show that cells generally migrate to the appropriate parts within the cortex, but Anderson says eventually the analysis should be more specific. “If you drive the cells to being visual cortex cells, do they selectively do they innervate the superior colliculus?”

The Sasai technique allows isolated embryonic stem cells to aggregate quickly in a culture media that does not contain serum. The balls of cells that subsequently form produce a high percentage of cortical precursors, and they even assemble into layers showing interactions of the type observed in the cerebral cortex: wave-like oscillations in calcium ions and regional pattern induction caused by signaling proteins Wnt and fibroblast growth factor.

Cell-surface markers indicate that these embryoid bodies even seem to form neurons according to a pattern dubbed ‘inside early, outside late’, in which outer layers of cells contain the neruons that form in later stages of development. Both human and mouse cells formed qualitatively similar structures, displaying four layers characteristic of the early cortex but not the two final, more mature layers. However, the human cells formed mushroom-shaped aggregates about ten times wider than those formed by the mouse cells. They also, as expected, took much longer to grow.

The organization and differentiation of the embryonic stem cells seems sensitive to cell density and culture media, and a different set of parameters may allow the formation of all six layers found in the neocortex. Sasai now plans to use real-time imaging to watch cell events and figure out what those parameters might be.

“This is the first time that such self-organization in cortical structures has been demonstrated for mammals,” says Luc Leyns, who studies embryonic patterning and cell differentiation at the Vrije Universteit Brussel in Belgium. He notes that while the spatial organization of the early embryonic cortex is nicely recapitulated in culture, the temporal sequence of the structures is different, and that this should be further investigated. Nonetheless, he sees several potential biomedical applications of Sasai’s work. If induced pluripotent stem cells created from particular patients were used, for instance, structures formed by people with and without neurodegenerative diseases could be compared. He also believes the assay could be adapted to screen out drugs with potential neurotoxic effects.

Sasai himself says he was both surprised and impressed by what he calls “the self-formation of a highly ordered pattern from patternless cells.” He thinks that this could lead to a shift in thinking for both research and medical applications; instead of focusing on producing certain cells, scientists could instead strive to make functional tissues.

Related articles

An intrinsic program of brain development


1. Gaspard, N. et al. An intrinsic mechanism of corticogenesis from embryonic stem cells. Nature 455, 351–357 (2008).

2. Eiraku, M. et al. Self-organized formation of polarized cortical tissues from ESCs and its active manipulation by extrinsic signals. Cell Stem Cell 3, 519–532 (2008).

Exogenous aids to reprogramming and self-renewal: Small molecules replace pluripotency gene and Wnt boosts self-renewal and reprogramming

A trio of recent papers shows that differentiated cells can be persuaded more easily into pluripotency or self-renewal by either adding small molecules or stimulating a signalling pathway found in most differentiating cells.

Following on from previous work in neural progenitor cells, researchers led by Sheng Ding of the Scripps Research Institute in La Jolla, California, were able to reprogram mouse embryonic fibroblasts (cultured skin cells) using only two of the standard four pluripotency genes1. By adding a small-molecule inhibitor of histone methyltransferase, BIX-01294, along with the calcium-channel agonist BayK8644, Ding and his colleagues found that they needed only to insert the genes for Klf4 and Oct4 to create induced pluripotent stem (iPS) cells that passed a stringent test for pluripotency — when mixed with mouse embryos, the iPS cells could contribute to sperm formation in the resulting mice. At around 0.2% efficiency, reprogramming rates are similar to those found using the standard four factors.

The researchers identified this combination through a series of phenotypic screens of around 2,000 small molecules. They added the molecules to fibroblast cultures, then looked for the formation of colonies typical of embryonic stem (ES) cells and for expression of the ES cell marker alkaline phosphatase. BIX-01294 had been previously recognized to boost reprogramming, presumably by affecting epigenetic regulation of gene expression. In fact, it is also able to generate IPS cells even without BayK8644, though not as effectively. The screen also found a DNA methyltransferase inhibitor called RG108 that enhanced reprogramming in combination with BIX, but because this mechanism of action is precedented, the researchers chose not to pursue it. However, the researchers describe the identification of the booster activity of BayK8644 as “intriguing”; it is known to induce intracellular signalling and the release of calcium ions. How exactly the two molecules might interact to make the presence of the transcription factor Sox2 unnecessary for reprogramming is a matter for further study.

Cell signalling is also involved in two other papers showing exogenous boosting of rates of reprogramming or self-renewal. A team led by Roel Nusse and Irving Weissman at Stanford University in California found that adding the signal protein Wnt to culture media can boost clonal outgrowth in populations of neural stem cells and that inhibiting Wnt signalling stalls attempts to clone or expand cell cultures2.

Another team led by Maria Pia Cosma at the Telethon Institute of Genetics and Medicine in Naples, Italy, found that periodically activating Wnt signalling boosts the ability of ES cells to reprogram genomes from specialized somatic cells, including neural stem cells and thymocytes3. The team fused the somatic cells with ES cells, and then increased Wnt signalling either by adding Wnt directly or by inhibiting the enzyme glycogen synthase kinase 3, which is inactivated in the course of normal Wnt signalling.

The fused somatic nuclei were demethylated in patterns resembling those found in ES cells, and also began expressing known pluripotency genes. Intriguingly, reprogramming rates corresponded to the accumulation of a specific threshold of beta-catenin, the intracellular mediator of Wnt signaling. Furthermore, the researchers only observed this effect when Wnt signalling was boosted in newly fused cells or in ES cells, not in differentiated cells. Thus, the researchers conclude that Wnt signalling somehow activates the reprogramming capacity of pluripotent nuclei.

[[Author Affiliation]] Monya Baker is editor of Nature Reports Stem Cells


1. Shi, Y. et al. Induction of pluripotent stem cells from mouse embryonic fibroblasts by Oct4 and Klf4 with small-molecule compounds. Cell Stem Cell 3, 568–574 (2008).

2. Kalani, M. Y.S. Wnt-mediated self-renewal of neural stem/progenitor cells. Proc. Natl Acad. Sci. USA 105, 16970–16975 (2008).

3. Lluis, F., Pedone, E., Pepe, S. & Cosma, M. P. Periodic activation of Wnt/-catenin signaling enhances somatic cell reprogramming mediated by cell fusion. Cell Stem Cell 3, 493–507 (2008).

Embryoid bodies get organized: Wnt signalling stimulates the beginnings of gastrulation

Left to their own devices, cultured pluripotent stem cells clump together. A hotchpotch of differentiated cells forms within these so-called ’embryoid bodies’, and in fact, the formation of embryoid bodies is a preliminary assay of good quality in embryonic stem (ES) cells and induced pluripotent stem (iPS) cells. Researchers led by Roel Nusse at Stanford School of Medicine in California have now observed something surprising in these much studied structures: they spontaneously begin a process called gastrulation, the cell movements that occur in mammalian embryos after implantation and which result in the formation of the three germ layers of the animal body.

“It is not a surprise to find that gastrulation can happen in embryoid bodies but the use of fluorescence-based tools allows the visualization of the gastrulation phenomenon occurring in vitro which mimics rather faithfully the in vivo process,” says Luc Leyns, who studies Wnt signaling and embryonic patterning at Vrije Universiteit Brussels in Belgium. Embryoid bodies generally contain single clusters of, say, neural or mesodermal cells, he says. “These results suggests that a coordinating mechanism exists but this one is still to be identified.”

The ‘gastrulation’ process in the embryoid bodies depends on a ubiquitous signalling protein known as Wnt, and though Wnt is known to be essential for gastrulation in mammalian embryos, no one had looked for it in embryoid bodies, says Nusse. His lab members Derk ten Berge and Wouter Koole used a series of established Wnt reporters to visualize Wnt signalling in embryoid bodies formed from mouse ES cells, and found a surprising level of organization1. The embryoid bodies began to develop regions that resembled the primitive streak, the area through which cells travel to form the mesoderm. This created a polarization within the embryoid bodies, and the formation of the primitive streak-like region depended on local activation of the Wnt-pathway. Once Wnt signaling starts in the embryoid body, it is self-reinforcing. Cells within this region underwent a switch known as the epithelial-to-mesenchymal transition and began to differentiated into mesoendodermal progenitors.

Wnt influences the differentiation in all sorts of cell types in many different ways, and the fact that Wnt signalling starts spontaneously in embryoid bodies suggests a range of techniques that might be used to control the differentiation of cultured stem cells, says Nusse. “If you can mimic that [guided differentiation] in cell culture, that is a step toward instructing ES cells to become committed to other fates.”

What’s unclear is what establishes Wnt signalling in the first place: some constituents of serum-based media seem to trigger it, so differentiation protocols that use media without serum might be more predictable, says Nusse.

Ethical oversight committees at universities do not allow researchers to perform experiments on human embryos or embryo-like entities after the first appearance of the primitive streak, which occurs at thebeginning of gastrulation. Nusse performed his experiments on mouse embryoid bodies, but he doubts that his findings will prevent work in human cells because embryoid bodies show only the very beginning of gastrulation. Nonetheless, he says, “What happens in an embryoid body is more like what happens in an embryo body than previously appreciated.”

1. ten Berge, D. et al. Wnt signaling mediates self-organization and axis formation in embryoid bodies. Cell Stem Cell 3, 508–518 (2008).


Comments are closed.