doi: 10.1242/10.1242/dev.00374
Delta-Notch signaling controls the generation of neurons/glia from neural stem cells in a stepwise process
Luc Grandbarbe1,
Julien Bouissac1,
Matt Rand2,
Martin Hrabé de Angelis3,
Spyros Artavanis-Tsakonas4,5 and
Eliane Mohier1,*
1 Laboratoire de Neurobiologie du Développement et de la
Régénération — CNRS, 5 rue Blaise Pascal, 67000
Strasbourg, France
2 Department of Anatomy and Neurobiology, College of Medicine, University of
Vermont, Burlington, VT 05405, USA
3 GSF, Institute for Mammalian Genetics, Ingolstaedter Landstrasse 1, D-85764
Neuherberg, Germany
4 Massachusetts General Hospital Cancer Center, Department of Cell Biology,
Harvard Medical School, Charlestown, Massachusetts 02129, USA
5 Laboratoire d'embryologie cellulaire et moléculaire —
Collège de France, 49B, Avenue de la Belle Gabrielle, 94736 Nogent sur
Marne, France
View larger version (19K):
[in a new window]
|
Fig. 1. Protocol used for analyzing the differentiation potential of neurospheres.
Fully dissociated neurospheres were grown in serum-free neurosphere culture
medium, containing EGF (20 ng/ml) for various times, generally not exceeding 3
days in order to minimize necrosis, which might affect the core of larger
spheres, and the generation of new stem cells, which might be at the origin of
a `subclone' whose developmental potential could interfere with
interpretation. At t=0, spheres (50-100) were deposited on coverslips
coated with polyornithine and allowed to differentiate in neurosphere culture
medium containing 2 ng/ml EGF, in order to reduce proliferation. After various
times of differentiation, spheres were fixed and processed for immunocytology,
and analyzed by confocal microscopy (the observation plane being at the basis
of the spheres where the cells differentiate).
|
|
View larger version (64K):
[in a new window]
|
Fig. 2. Compared differentiation potential of wild-type and
Dll1lacZ/lacZ mutant neurospheres. (A) Kinetics of
differentiation were analyzed for wild-type (I, III, V, VII) and
Dll1lacZ (II, IV, VI, VIII) spheres. Three-day-old spheres
were fixed after 5 hours (I, II), 10 hours (III, IV), 48 hours (V, VI) and 6
days (VII, VIII) on polyornithine. Markers used for immunostaining were:
anti-MAP2 (I-VI) coupled to Cy3, for neurons (red); anti-PDGFR (I-IV) coupled
to Alexa 488, for OPCs (green); anti-GFAP (V, VI) coupled to Alexa 488, for
astrocytes (green); and anti-O4 coupled to Cy2 (green) (VII, VIII). In all
cases, nuclei were visualized by TOPRO (blue). (B) Approximate quantification
of the results. Each of the cell type was quantified at various
differentiation times: MAP2-positive cells after 48 hours; GFAP-positive cells
after 48 hours; oligodendrocytes were estimated from PDGFR-positive cells
after 10 hours on polyornithine. The data were cumulated and expressed as
percentages of total cell number estimated from TO-PRO staining. Scale bar: 50
µm.
|
|
View larger version (65K):
[in a new window]
|
Fig. 3. Neurospheres differentiate in response to Notch signaling with different
sensitivity for neurons and astrocytes. (A) Partial inactivation of Notch
signaling in Dll1lacZ/+ heterozygous spheres induces a
moderate increase in neurons (MAP2, in red), a decrease in astrocytes (GFAP,
in green) and dramatically alters their morphology (after 3 days of
differentiation). (B) Quantitative estimation for A. Data are representative
of three independent experiments. (C-E) Differentiation of
Dll1lacZ/lacZ homozygous spheres in response to various
concentrations of J1EC. (C) Experimental protocol:
Dll1lacZ/lacZ mutant spheres were allowed to differentiate
in the presence of various concentrations of J1EC. (D) Triple
immunostaining using anti-MAP2 (red), anti-GFAP (green) and TO-PRO (blue). (E)
Quantitative results of D. Data are representative of three independent
experiments. Scale bar: 50 µm. I,II,III and IV indicate the dilutions of
J1EC.
|
|
View larger version (61K):
[in a new window]
|
Fig. 4. Effect of time-dependant activation of Notch in
Dll1lacZ/lacZ mutant spheres on neurons and astrocytes.
Three-day-old Dll1lacZ/lacZ mutant spheres were incubated
in the presence of J1EC for various time intervals. as described in
the schematic protocol (A). (B) Immunocytological analysis. Spheres were
immunostained with antibodies against MAP2 (red), GFAP (green). (C)
Quantitative estimation was as described in experimental procedures. Data are
representative of three independent experiments. Scale bar: 50 µm. I,II,III
and IV indicate the addition of J1EC, as defined in A.
|
|
View larger version (63K):
[in a new window]
|
Fig. 5. Effect of Notch activation on astrocyte differentiation in
Dll1lacZ/lacZ and wild-type spheres. (B) Two-day-old
Dll1lacZ/lacZ mutant (I, II, III) or wild-type (IV, V, VI)
spheres were exposed to J1EC for various time intervals as
described in A. (B) Triple immunostaining involved anti-MAP2 (red), anti-GFAP
(green) and TO-PRO (blue). Colocalized markers are rarely observed, and are
likely to result from overlapping cells as they appear in fields that are
particularly dense, and probably biologically not significant. Therefore, we
consider the yellow signals (V, VI, arrows) as the presence of neurons (red)
overlapping astrocytes (green). (C) Data are representative of two independent
experiments. Scale bar: 50 µm.
|
|
View larger version (64K):
[in a new window]
|
Fig. 6. Effect of time-dependant activation of Notch in
Dll1lacZ/lacZ mutant spheres on the production of OPCs (B,
I-III) and on oligodendrocytes (B, IV-VI). (A) Experimental protocol: 24 hours
(left panel) or three-day-old (right panel) Dll1lacZ/lacZ
mutant spheres were incubated in the presence of J1EC for various
time intervals. (B) Spheres were immunostained with anti-MAP2 (red) and
anti-PDGFR (green) for OPCs production (I-III); anti-04 (green) for
oligodendrocytes production (IV-VI). (C) Quantitative estimations of neurons
and OPCs (from I-III) was as described in experimental procedures. Data are
representative of three independent experiments. Scale bar: 50 µm.
|
|
View larger version (20K):
[in a new window]
|
Fig. 7. Tentative model for the role of Notch in the generation of neurons/glia
from neural stem cells in neurospheres. An initial EGF-responsive neural stem
cell (NSC1) asymmetrically divides, giving rise to a second stem cell (NSC2)
and a progenitor (P1) that appears as inevitably fated to a neuronal identity.
As a neuronal precursor, this cell is endowed with a limited proliferation
capacity and is responsible for the few neurons generated under all
circumstances. The asymmetrical division of NSC2 generates a second precursor
(P2). The activation of Notch by P1-produced Dll1 prevents P2 from adopting a
neuronal fate. Instead, P2 becomes irreversibly committed to a glial fate. The
model postulates that P2 has the potential to acquire either the astrocytic or
the oligodendroglial identity through a mechanism independent of Notch
signaling. In a second step, Notch would affect the differentiation decision
of the precursors already committed to a neuronal or a glial lineage. It would
inhibit the differentiation of neurons and oligodendrocytes, while promoting
the differentiation of astrocytes.
|
|
© The Company of Biologists Ltd 2003
