The idea that some of the genes at the root of familial Alzheimer disease are involved in neurogenesis just got a little boost. In this week’s PNAS online, Pasko Rakic and colleagues at Yale University, New Haven, Connecticut, report that Notch signaling is crucial for determining the fate of newborn neurons in the postnatal mouse dentate gyrus, a birthplace for neurons. Notch signaling depends on its cleavage by γ-secretase, whose catalytic component is presenilin, and for this reason the finding suggests that presenilins may play a crucial role in the fate of neural progenitors. This is worth considering in the context of Alzheimer disease, given the role of presenilin (PS) mutations in familial Alzheimer disease and the interest in PS inhibitors as a potential treatment. In other neurogenesis news, Mark Henkemeyer and colleagues at the University of Texas Southwestern Medical Center, Dallas, report that ephrin B receptors are needed for proper adult neurogenesis in mice and suggest that targeting ephrin signaling could be one approach to ramping up production and integration of new neurons in old brains.

Since adult neurogenesis was discovered as a mammalian phenomenon, the idea of promoting newborn neurons for therapeutic purposes has attracted researchers. The vision of coaxing endogenous new neurons to take over from damaged or dying brethren is irresistible, particularly for neurodegenerative disorders such as Parkinson or Alzheimer disease. However, as Rakic and colleagues write, our knowledge of the molecular mediators that permit neurogenesis is in its infancy.

These two papers take a step toward filling that knowledge gap. Finding that Notch1 is robustly expressed in hotbeds of adult neurogenesis—the mouse subependymal zone and subgranular zone of the dentate gyrus—Rakic and colleagues evaluated the role of the signaling receptor in neuronal differentiation. First author Joshua Breunig and colleagues examined cell fates in inducible Notch1 knockouts and also in inducible models overexpressing the Notch intracellular domain (NICD). They found that if the Notch1 gene was knocked out during postnatal days 10 to 14, then 1 week later, the number of proliferating cells in dentate gyrus was down, while the number of cells exiting the cell cycle almost doubled. (Cells were counted as proliferating when they expressed the cell cycle marker Ki67, and as exiting the cell cycle when they were positive for the DNA synthesis marker iododeoxyuridine, but negative for Ki67.) In contrast, induced overexpression of NICD led to a threefold increase in proliferating cells in the dentate gyrus and about a fourfold decrease in SGZ cells exiting the cell cycle. All told, the experiments indicate that Notch1 signaling promotes proliferation of cells in the hippocampus, the authors write. The authors did not report performing this experiment in older mice.

The researchers recapitulated the conditional knockout effect by using γ-secretase inhibitors, supporting the idea that proliferation requires the processing of Notch1. On that note, the researchers found that nuclear NICD staining is stronger in more mature neurons (NeuN- and doublecortin-positive) and interpret this observation to mean that Notch signaling may also be important for the maturation of newborn neurons.

If abolishing Notch1 drives progenitor cells out of the cell cycle, what becomes of them? Ironically, it appears they become neurons. When Breunig and colleagues examined subgranular zone and granule cell layers 1 week after knocking it out, they found that more cells in Notch1-free tissue than in control tissue were classified as neuronal. In contrast, induction of NICD decimated neuron numbers. The same reversal occurred in adult (4-6-month) mice, too, though the effect was smaller.

Rakic and colleagues propose that Notch1 works like a binary switch. When off, proliferation of progenitors shuts down and most cells become neurons. When on, proliferation ramps up, but most cells become glia. Thus, “the level of Notch1 regulates the magnitude of neurogenesis from postnatal progenitor cells,” they write. It also has another effect. Breunig showed that dendritic arborization of new neurons is far greater 23 days after induction of NICD overexpression, but stunted after conditional knockout of Notch1.

As Notch signaling requires presenilin, this report bolsters earlier reports that the protease may be needed for adult neurogenesis. For example, Paul Wen and colleagues at Mt. Sinai School of Medicine in New York City reported that presenilin FAD mutations crimp adult neurogenesis (see ARF related news story), while Gabriel Corfas and colleagues at Harvard Medical School showed that cleavage of the Erb4 receptor tyrosine kinase by presenilin helps progenitors assume a neuronal fate (see ARF related news story).

Tyrosine kinases also feature in the second paper. Ephrins and the tyrosine kinase ephrin receptors (Ephs) have been linked to proliferation, but not to hippocampal progenitor cells. Henkemeyer and colleagues report that EphB1 expression is robust in the SGZ. By crossing EphB1-negative mice with animals expressing enhanced green fluorescent protein (eGFP) under control of the nestin promoter, first author Michael Chumley and colleagues found that loss of EphB1 halved the number of eGFP-positive cells in the corona of the dentate gyrus of 8-10-week-old mice. Curiously, eGPF-positive cell numbers were also about half in aged mice, too, suggesting that loss of EphB1 does not abolish cell proliferation completely. Instead, the cells loss may be due to faulty migration, the authors conclude.

The researchers found eGFP progenitors scattered throughout the hippocampus, rather than concentrated at the boundary between the subgranular zone and the granule layer, where they belong. Similarly, they found that numbers of proliferating cells were halved in the SGZ and that dividing cells (BrdU-positive) turned up in the granule layer, the hilus, and the molecular layer. Most of these proliferating cells are fated to become neurons as they also express doublecortin.

Besides EphB1, other such receptors appear to regulate the fate of new neurons, as well. When the authors examined EphB1/EphB2 double knockouts, they found dramatic morphological changes in the hippocampus that are not apparent in EphB1 knockouts alone, suggesting that EphB2 may compensate for its sibling. As for the ligand, the authors found that Ephrin-B3 knockouts recapitulate many of the phenotypes of EphB1 knockouts and that Ephrin-B3 is highly expressed in the SGZ, suggesting that it may be the major ephrin ligand controlling neurogenesis.

How might ephrin-EphB signaling control migration of progenitor cells? One possibility is that the protein is important for cell polarity. The authors found that dendritic branches, which normally extend radially toward the molecular layers once cells have projected through the granule layer, appear to sprout prematurely and randomly in EphB1-negative brain. “In total, the premature dendritic branching and presence of abnormal, apparently randomized cellular projections indicate that loss of EphB1 expression leads to severe disruption of the normal growth and polarity of cellular processes extending from hippocampal stem/progenitor cells,” they write.—Tom Fagan


  1. The paper by Breunig et al. represents an elegant set of experiments, performed by a talented young investigator and aimed at dissecting the molecular mechanisms responsible for postnatal neurogenesis. Using tamoxofen-inducible Notch1 knockout mice and inducible Notch intracellular domain (NICD) transgenic mice, the authors show that Notch acts as a molecular “cell fate trigger” for determining the proliferation versus differentiation decision. Specifically, loss of Notch1 caused progenitors to exit the cell cycle while conversely, overexpression of the constitutively active NICD domain caused a striking three- to fourfold increase in proliferating cells in the dentate gyrus and subgranular zone. Importantly, the investigators validated their findings in this system by taking a pharmacological approach, where they used γ-secretase inhibition to recapitulate the effects observed in the inducible Notch1 knockout mice.

    These findings have relevance for Alzheimer disease (AD) at a number of levels. The guiding principle for γ-secretase inhibition as an AD therapeutic approach has been the amyloid cascade hypothesis, which in its simplest form puts forth that cerebral amyloidosis produces a series of pathological downstream events that perpetrate AD. However, the results of Breunig et al. lead us to question whether such a therapeutic avenue will, in principle, be beneficial. This is because reducing amyloidosis by γ-secretase inhibition may come at the cost of shutting down adult neurogenesis occurring in the brains of AD patients. That is, of course, assuming that adult neurogenesis is even beneficial for reducing AD severity, a premise that remains questionable.

    These results also cause us to critically consider the intersection between Notch and amyloid precursor protein (APP), which, as the above editorial rightly points out, both have γ-secretase cleavage in common. Does γ-secretase indiscriminately cleave both molecules, or does it perhaps give “preferential treatment” to one or the other? The latter has been suggested by the work of Paul Greengard’s group, who have shown that Gleevec is able to reduce APP cleavage/Aβ generation while leaving Notch cleavage unperturbed (Netzer et al., 2003). An answer to this question may allow for a more tailored AD therapeutic strategy aimed at reducing cerebral amyloidosis without the potentially unwanted side effect of shutting down adult neurogenesis.


    . Notch regulates cell fate and dendrite morphology of newborn neurons in the postnatal dentate gyrus. Proc Natl Acad Sci U S A. 2007 Dec 18;104(51):20558-63. PubMed.

    . Gleevec inhibits beta-amyloid production but not Notch cleavage. Proc Natl Acad Sci U S A. 2003 Oct 14;100(21):12444-9. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Familial Alzheimer's Presenilin Gene Perturbs Neurogenesis?
  2. Squelching Gene Expression Controls Development, Neurodegeneration

Further Reading

Primary Papers

  1. . EphB receptors regulate stem/progenitor cell proliferation, migration, and polarity during hippocampal neurogenesis. J Neurosci. 2007 Dec 5;27(49):13481-90. PubMed.
  2. . Notch regulates cell fate and dendrite morphology of newborn neurons in the postnatal dentate gyrus. Proc Natl Acad Sci U S A. 2007 Dec 18;104(51):20558-63. PubMed.