In last week's PNAS online, researchers reported that murine embryonic stem cells exposed to pyrrolpyrimidines differentiate into neurons. This result raises hope that the fate of stem cells in vivo may one day be controlled by the use of small molecules.

Embryonic stem cells can be coaxed to differentiate into a variety of cell types, such as muscle, glia, and neurons. Traditionally, this has required that the cells be exposed to specific proteinaceous growth factors or cytokines (see ARF related news story), which are difficult to administer in vivo. While the small, cell-permeable retinoic acid has also been used to provoke differentiation, this pleiotropic compound causes the cells to differentiate into many different cell types.

Principal author Peter Schultz and colleagues from The Scripps Research Institute, La Jolla, and the Novartis Research Foundation, San Diego, California, set about to identify small molecules that could lead stem cells specifically down the path to full-fledged neurons. First author Sheng Ding and coworkers used genetic engineering to equip mouse stem cells with a luciferase reporter gene that is under the control of the neuronal tubulin gene promoter. Because this promoter is only active in neurons, the authors could use luciferase activity to identify neuronal phenotypes among the stem cells and their progeny. Using high-throughput screens to expose these modified stem cells to thousands of small molecules, Ding and colleagues found chemicals that promoted neuronal differentiation.

From this preliminary analysis, the authors chose to focus on pyrrolpyrimidines. By tweaking the chemical substituents attached to the pyrrolpyrimidine backbone, the authors found a compound, dubbed TWS119 that could induce neuronal differentiation in 40 to 60 percent of the stem cell population. In these cells, luciferase activity was accompanied by positive staining for a variety of neuronal cell markers, including TuJ1, NeuN, glutamate, and neurofilament-M. The authors also report that TuJ1-negative cells tested negative for muscle and glial cell markers, but did test positive for nestin, a sign that the cells are neuronal progenitors.

Curious to know how TWS119 works, Ding and colleagues used an affinity matrix to isolate stem cell proteins that bind to the compound. Only two proteins bound with high affinity, and the authors used mass spectroscopy to show that these are both isoforms of glycogen synthase kinase-3β (GSK3β). This kinase is known to be involved in the "wingless" signaling pathway, phosphorylating the pathway component β-catenin and stimulating its degradation. When Ding and colleagues examined TWS119-treated pluripotent cells, they found that levels of β-catenin had increased over 10-fold, consistent with inhibition of GSK3β. β-catenin is already linked to neurogenesis (see Walsh section in ARF related news story) and to presenilin (see de Strooper section in ARF related news story). It also is worth noting that GSK3β has been implicated in the phosphorylation of the protein tau, which is found in the neurofibrillary tangles that often occur with Alzheimer's disease (see ARF related news story).
“This is a very interesting idea to direct cell lineage,” commented Kiminobu Sugaya, University of Chicago, “though I read this paper with a little bit of surprise because regulation of only one enzyme, GSK3β, seems sufficient to lead embryonic stem cells to neurons. I would like to see further investigation of the mechanism of action of this small compound” he added.—Tom Fagan


  1. This paper reports interesting data on the handles that might control the differentiation and phenotype of stem cells. The authors used a high-throughput phenotypic screen to identify a small synthetic molecule called TWS119, which induced transformation of a considerable percentage of mouse embryonic stem cells into neurons. Upon further characterization of this compound, the authors demonstrated that this molecule is a pyrrolpyrimidine and that its main target is glycogen synthase kinase (GSK)3β, a well-known kinase in Alzheimer disease.

    In various neurodegenerative diseases, the loss of distinct neuronal populations causes severe neurological symptoms. Hence, the possibility of replacing these lost cells and integrating them again in existing neuronal circuits is an attractive, yet complex therapeutic avenue. In addition, several studies have now shown that in the adult and even aging brain, neurogenesis continues to occur, albeit limited in number and only in two locations, i.e., the subventricular zone and the hippocampus (Heine et al., 2003, in press). In other brain areas like the cortex, adult proliferation occurs, as well, but here it gives rise to glia cells. For some reason, adult neurogenesis is not possible in other brain areas.

    Although relatively little is known to date about the factors that control—and within the brain, in fact prevent—the differentiation of newborn cells into a neuronal phenotype, another attractive strategy would be to "recruit" the endogenous potential of the brain's stem cells for repair in areas other than the hippocampus. In my view, the currently described molecules are very interesting candidates for such approaches.

    Stem cell differentiation clearly depends on the surrounding conditions, and in the present study, indeed, was limited to selected cell populations, i.e., P19 cells and primary mouse ESCs. Furthermore, treatment with 1 μM TWS119 caused about 30-40 percent of the cells to differentiate in neurons, whereas, surprisingly, transient treatment for two days followed by two more days of compound-free incubation resulted in even higher (40-60) percentages of neurons. Apparently, extended exposure of neural progenitors to early differentiation signals affects late-stage neuronal maturation. Moreover, in primary mouse ESCs, treatment with an even lower dose caused neuronal differentiation in 50-60 percent of the cells. Clearly, this molecule must be further tested to determine whether it can exert similar effects in other cells or cell lines and model systems.

    The finding that TWS119 is a target for GSK3b is intriguing. At the same time, many proteins are phosphorylated by GSK3, so it could be rather unspecific. GSK3 is a well-conserved, generally active kinase that modifies the function of a variety of proteins. It is one of the candidates for producing excessive phosphorylation of tau, which may harm or even protect neurons in Alzheimer disease (Spittaels et al., 2000). In addition to the affinity studies and the suggestions of a possible role in the Wnt pathway or the control of bHLH transcription factors, further functional and biochemical characterization is needed before this TWS molecule can be further implicated in tau phosphorylation. As in other studies, such a role would be consistent with the suggested close interplay between neuronal (de)differentiation processes, cytoskeletal plasticity and tau alterations, processes that occur both in early development as well as in neurodegenerative conditions like Alzheimer¡¦s disease.

    Overall, this paper is one of the first to identify possible (synthetic) candidate molecules that could, together with existing biological compounds, bear relevance for triggering (endogeneous) stem cell division and differentiation into a neuronal phenotype. Clearly, more information on TWS119 regulation and the uniformity of its working mechanism is needed, in addition to the important question of whether similar effects can be obtained upon local application in in-vivo approaches.


    . Glycogen synthase kinase-3beta phosphorylates protein tau and rescues the axonopathy in the central nervous system of human four-repeat tau transgenic mice. J Biol Chem. 2000 Dec 29;275(52):41340-9. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Smells Like Neuronal Proliferation!
  2. Not Strictly AD—Selected Basic Science Highlights from a Press Seminar
  3. Notes from International AD/PD Conference 2003 in Seville
  4. Wingless Pathway Helps Tauopathy Take Off

Further Reading


  1. . Molecular manipulation of neural precursors in situ: induction of adult cortical neurogenesis. Exp Gerontol. 2003 Jan-Feb;38(1-2):173-82. PubMed.

Primary Papers

  1. . Synthetic small molecules that control stem cell fate. Proc Natl Acad Sci U S A. 2003 Jun 24;100(13):7632-7. PubMed.