Search tips
Search criteria 


Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. 2010 April; 30(8): 1946–1957.
Published online 2010 February 12. doi:  10.1128/MCB.01419-09
PMCID: PMC2849467

Notch Exhibits Ligand Bias and Maneuvers Stage-Specific Steering of Neural Differentiation in Embryonic Stem Cells[down-pointing small open triangle]


Notch dictates multiple developmental events, including stem cell maintenance and differentiation, through intercellular communication. However, its temporal influence during early development and, of particular interest, its regulation of binary fate decision at different stages during neurogenesis are among the least explored. Here, using an embryonic stem cell (ESC) model, we have deciphered Notch ligand preference during ESC commitment to different germ layers and determined the stage-specific temporal effect of Notch during neural differentiation. ESCs during maintenance remain impervious to Notch inhibition. However, Notch activation promotes differentiation even in the presence of leukemia inhibitory factor (LIF), displaying ligand preference-associated lineage discrimination, where Jagged-1 favors neural commitment and Delta-like-4 favors the mesoderm. This differential ligand action involves a combination of Notch receptors influencing specific downstream target gene expression. Though Notch activation during early neural differentiation specifically promotes neural stem cells or early neural progenitors and delays their maturation, its inhibition promotes late neural progenitors and expedites neurogenesis, with a preference for neurons over glia. However, gliogenesis is promoted upon Notch activation only when executed in combination with ciliary neurotrophic factor. Thus, our investigation underscores a multifaceted role of Notch, demonstrating the interdependency of ligand usage and lineage specification and Notch acting as a master switch, displaying stage-specific influence on neurogenesis.

Notch exerts its pivotal influence by signaling through ligand-receptor interaction occurring between the neighboring cells. It regulates several vital processes during early embryogenesis and organ development (2, 6). Although originally identified as a neurogenic gene regulator in Drosophila melanogaster, its involvement is not restricted to neural development alone (2). A single Notch receptor, along with its sole ligand, Delta, is known to orchestrate multiple functions in Drosophila. However, it has evolved into four Notch receptors (Notch1 to -4) and five ligands (Dll1, Dll3, Dll4, Jag1, and Jag2) in mammals, reflecting the complexities involved in regulating the associated functions (40). In fact, an understanding of the emergence of diverse ligands and receptors despite the existence of their functional redundancy remains elusive. While the basic signaling mechanism is evolutionarily conserved, the ligands and receptors have been elucidated to bear independent functions. Hence, the definitive ligand-receptor dependency or their interplay in specific functional outcomes remains to be ascertained.

Notch has been reported to induce generation of heterogeneous cell types from a developmentally similar population of cells (2). Accordingly, its involvement has been illustrated in the regulation of a number of processes in adult stem cells (34). In hematopoietic and neural stem cells (NSCs), the role of Notch in maintenance, fate determination, proliferation, and survival has been extensively studied (8, 34). However, its definitive role during embryonic stem cell (ESC) maintenance and differentiation into various lineages remains to be explored. Indeed, understanding the basic biology of ESCs with respect to the signaling cues underlying their maintenance and cell fate determination is critical to harnessing their true potential. In ESCs, Notch has been reported to promote neural commitment (23), supposedly a default process (38). However, neural maturation depends on the seeding density (21), thus reflecting a possible involvement of Notch signaling during this process.

Notch signaling commences upon binding of its ligand to the receptor, which undergoes cleavage by γ-secretase (gS) enzyme complex in the membrane to release its intracellular domain (NICD). NICD directly enters the nucleus and regulates the transcriptional activities of several target genes (1, 14, 36). Hence, canonical Notch signaling in cells can be attenuated by inhibiting gS enzyme complex. Conversely, the signaling can be activated by ectopic ligand binding to the receptor (35). However, the ligand-receptor interactions seem to be quite divergent depending on the cell type, and so are the genes targeted upon Notch activation. Therefore, a quite complex receptor and ligand interaction has been speculated in the regulation of the signaling pathway.

Here we have demonstrated a fascinating paradigm of Notch ligand bias influencing germ layer commitment, using an ESC model. Despite being cultured under maintenance conditions, ESCs undergo differentiation upon Notch activation, displaying an interesting differential ligand response, where Jagged-1 (Jag1) promotes neural commitment and Delta like-4 (Dll4) promotes that for the mesoderm. Moreover, Notch exerts differential and stage-specific temporal influence during generation of neural progenitors and their differentiation into neurons and glia. Overall, our study underscores a crucial influence of Notch during cell fate determination, manipulation of which will facilitate directed differentiation of ESCs into cells of interest in vitro.


ESCs' maintenance and differentiation.

All experiments were carried out using a murine ESC line, D3 (ATCC, Manassas, VA), and the derived nes-EGFP stable clone (21). ESC maintenance and differentiation were carried out as described elsewhere (20). In brief, ESCs were maintained in culture using Dulbecco's modified Eagle medium (DMEM) with leukemia inhibitory factor (LIF) (1,000 U/ml; Chemicon, Billerica, MA), l-glutamine, penicillin-streptomycin, nonessential amino acids (all from Invitrogen, Grand Island, NY), and β-mercaptoethanol (Sigma-Aldrich, St. Louis, MO) and were passaged every 48 h. Neural differentiation was carried out using the monoculture strategy as described previously (21), and the growth characteristics and differentiation were monitored under the microscope (TE2000U; Nikon, Japan) and were documented using a DXM-1200 camera and ACT-1 software (Nikon, Japan). To assess the role of Notch during ESC maintenance and differentiation, both Notch activation (39) and inhibition (37) conditions were attempted. For Notch activation, recombinant proteins, mDll4 and rJag1 (R&D systems, Minneapolis, MN), were either immobilized on tissue culture dishes, followed by cell seeding on those dishes, or added as soluble proteins for detection of their temporal influence during neurogenesis, as stated. In fact, immobilized ligands displayed a comparatively better response in terms of their Notch activation potential in ESCs than by the direct addition to the medium (see Fig. S1E in the supplemental material). For Notch inhibition, a gS inhibitor (gSI), L-685458 (Calbiochem, La Jolla, CA), was used at a concentration of 4 μM in all experiments unless otherwise specified.

RNA isolation and transcript analysis.

Total RNA was isolated using TriReagent (Sigma), following the manufacturer's protocol, and intact RNA samples were processed further for reverse transcription-PCR (20). The primer sequences can be made available upon request. Quantitative PCR (qPCR) was performed as described previously (21) in an iQ5 thermocycler (Bio-Rad, New South Wales, Australia) using SYBR green supermix (Bio-Rad). Relative expression levels of each gene were analyzed, normalizing to β-actin transcription, and the fold difference was calculated, keeping the values in the control with LIF (+LIF) at 1.


The cellular phenotypes were studied by immunocytochemistry, following the standard protocol (21). In brief, after fixing of the cells with 4% paraformaldehyde, cells were permeabilized (if necessary) using 0.025% Triton X-100 (Sigma) and blocked with 5% fetal bovine serum (FBS) in phosphate-buffered saline (PBS). The cells were incubated with either of the primary antibodies anti-glial fibrillary acidic protein (anti-GFAP; Sigma); anti-microtubule-associated protein 2 (anti-MAP2), anti-activated Notch (anti-aNotch), and anti-Numb (all from Abcam, Cambridge, MA); and anti-Oct4 and anti-Nanog (Santa Cruz, Santa Cruz, CA), followed by incubation with Cy3- or Cy2-conjugated (Chemicon) secondary antibodies for fluorescent labeling. 4′,6-Diamidino-2-phenylindole (DAPI) (Sigma) was used as a nuclear counterstain. In each case, the negative control was performed with the replacement of respective primary antibodies with FBS. The slides were observed under a laser scanning confocal microscope (TE2000E; Nikon, Japan) to detect enhanced green fluorescent protein (EGFP), as well as Cy3/Cy2 labeling. The numbers of differentiating neurons and glia were scored by counting (20×) the MAP2+ and GFAP+ cells among the DAPI+ cells by considering 6 to 8 random fields/sample per set. For calculation of neural arborization, images were processed using the ImageJ software program (NIH, Bethesda, Maryland), converting pixel units to appropriate μm lengths depending on the image magnification.

Flow cytometry quantification.

For flow cytometry quantification, a single-cell suspension was prepared and the cells were incubated with primary antibodies for 30 min, followed by washing and incubation with secondary antibody for 15 min. Both the primary and secondary antibodies were suspended in staining solution containing 2% FBS and 0.05% saponin in PBS, and the incubations were carried out at 4°C. Immunostaining of the single-cell populations was performed using primary antibodies: anti-Sox1 (Chemicon), anti-Nestin (Rat-401; Developmental Studies Hybridoma Bank, Iowa City, Iowa), and anti-TuJ1 (Sigma), followed by phycoerythrin (PE)-conjugated secondary antibodies (BD Biosciences, San Jose, CA) for fluorescent labeling. Quantification was carried out using a FACSCalibur instrument (Becton Dickinson, Singapore) equipped with a 488-nm argon-ion laser (15 mW). About 10,000 viable cells were analyzed per sample, following acquisition settings with appropriate controls. The emitted fluorescence of EGFP and PE were measured in log scale at 530 nm and 585 nm, respectively. The analysis was performed using the CellQuestPro software program (Becton Dickinson).

Luciferase assay.

ESCs were transfected with the 12×CSL-luc vector (a kind gift from Urban Lendahl) (3 μg) by nucleofection (Lonza nucleofection kit) and were plated with various treatments, as appropriate. Cells were harvested at 24 and 48 h posttransfection, and luciferase activity was measured by using the LucLite luciferase assay kit (Perkin Elmer). Since ESCs also demonstrated luciferase activity without exogenous Notch activation, indicating an endogenous Notch response, fold inductions were calculated relative to levels for the untransfected samples. However, the statistical significance was calculated by comparison with the respective controls.

Statistical analysis.

All data were presented as means ± standard errors of the means (SEM) from 4 to 10 experiments, and statistical significance was calculated using Student's paired t test (Sigmaplot, San Jose, CA). P values were calculated in comparison with the control and were represented as follows: *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.


Notch is dispensable for ESC maintenance.

Active canonical Notch signaling in cells is determined by the release of cleaved NICD, also known as activated Notch (aNotch), from the membrane receptor, followed by its nuclear localization. Accordingly, the status of this signaling pathway was assessed in ESCs by monitoring the localization of aNotch. As seen in Fig. Fig.1A,1A, undifferentiated ESCs, when cultured either in the presence or absence of LIF (+LIF or −LIF, respectively), displayed aNotch in the nucleus (1, 14), and that could be attenuated by using gSI (37). Hence, Notch signaling was deduced to be operational in ESCs. Further, to decode the connotation of this signaling, ESCs were exposed to gSI in culture under both +LIF and −LIF conditions. Under the +LIF condition, with sustained Notch inhibition for 6 days, no apparent alteration in cellular morphology and characteristics was seen (Fig. (Fig.1B);1B); rather, a marginal increment in cell number was observed. Conversely, inhibition under the −LIF condition led to the formation of cells with a large, flattened shape. Expression of the pluripotency markers Nanog (5, 25) and Oct4 (29) also remained unaltered under the +LIF condition, while the same decreased for the −LIF condition irrespective of Notch inhibition (Fig. (Fig.1C;1C; see also Fig. S1A to D in the supplemental material). Moreover, Numb, the Notch antagonist, expressed in asymmetrically dividing cells during differentiation (11), was not detected in cells when inhibited under the +LIF condition. However, a robust increase in its expression was detected under the −LIF condition just after 1 day of inhibition (Fig. (Fig.1D).1D). Collectively, these results implied that the consequence of Notch inhibition did not override the effect of LIF, and therefore, ESCs retained their inherent characteristics during maintenance. Hence, Notch signaling was speculated to be dispensable for ESC maintenance.

FIG. 1.
Notch involvement during ESC maintenance. (A) Notch inhibition attenuated nuclear localization of activated Notch (aNotch) in ESCs. The effect was more pronounced for −LIF than for +LIF, as seen after 1 day of inhibition. (B) ESC morphology ...

Notch-induced differentiation during ESC maintenance.

The default occurrence of Notch signaling in ESCs despite no apparent role in their maintenance poses a conundrum. To solve this paradox, ESCs were subjected to Notch signaling induction by exposing them to exogenous ligands either by culturing them over immobilized ligands or by using the ligands in soluble form. The ligands that activate Notch signaling in vertebrates are categorized as either Delta-like or Serrate-like based on the number of epidermal growth factor (EGF)-like repeats present in the N terminus (extracellular domain) of these ligands. These subtypes also differ in their intracellular domain. We chose the recombinant proteins for one member from each group (Dll4 and Jag1, respectively) that was available as a representative ligand subtype to activate Notch signaling in our experiments. The overexpression or genetic manipulation of NICDs was not preferred, since this could not have followed the intrinsic ligand-mediated signaling in the system per se. Moreover, our approach avoided bias among four NICDs for overexpression. The induction of Notch signaling under the stated conditions was authenticated by monitoring the luciferase activity following transfection of ESCs with 12×CSL-luc vector. As seen in Fig. S1E in the supplemental material, both Jag1 and Dll4 enhanced reporter activity, while gSI inhibited the same. Interestingly, when Notch signaling was induced in ESCs by culturing them over immobilized ligands, Dll4 and Jag1 (39), it resulted in ESC differentiation even under the +LIF condition. Cells lost their characteristic undifferentiated morphology (Fig. (Fig.1E),1E), accompanied by a significant decrease in Oct4 expression (Fig. (Fig.1F),1F), indicating the differentiated status of ESCs. However, the quantitative analysis of Nanog transcripts in the differentiating population in the Jag1-treated group revealed a dose-dependent contrasting response. Though there was a significant increase in Nanog expression with a lower Jag1 concentration (1 μg), it decreased at a higher concentration (10 μg) (Fig. (Fig.1G).1G). Unlike Jag1, Dll4 influenced a dose-independent decline in Nanog expression (Fig. (Fig.1G).1G). Similarly, immunocytochemical detection revealed a decrease in Nanog expression upon Notch activation except with lower Jag1 levels in the +LIF condition, where the difference remained relatively indiscernible from that of the control (see Fig. S1D). Oct4 expression, however, corroborated well with its transcription profile and remained low upon Jag1 (1 μg) treatment under the +LIF condition compared to that for the control (see Fig. S1F), thus suggesting a prodifferentiating influence of Notch on ESCs. Interestingly, Nanog staining appeared relatively brighter upon Notch inhibition than that for the vehicle control (see Fig. S1D). Hence, together with the inhibition studies, ESC maintenance overall entailed the possibility of a basal level of Notch signaling; though attenuation of this did not influence its undifferentiated state to an appreciable extent, its augmentation certainly led to differentiation.

Lineage commitment upon Notch activation.

To further assess the Notch induced-differentiation status in ESCs, whether spontaneous or to a specific lineage, expression of early-commitment genes was analyzed. Fgf5 (12), the primitive ectoderm marker, which normally appears during early commitment, was detected at a quite negligible level in ESCs maintained for 6 days with LIF. However, in the presence of ligands, it exhibited a disparate expression profile, thereby reflecting the subsistence of divergent signaling mechanisms executed by Notch (Fig. (Fig.2A).2A). While Dll4 treatment decreased Fgf5 expression under both +LIF and −LIF conditions, Jag1 influenced its dose-dependent increase under the +LIF condition and with no difference under the −LIF condition from that for respective controls. Further, to deduce the lineage identity, expression of early germ layer-specific genes, such as Nestin (Nes) (ectoderm), Brachyury (Bry) (mesoderm), and α-feto protein (Afp) (endoderm) was assessed. For +LIF, Dll4 treatment did not influence any particular lineage specification and instead inhibited ectoderm and endoderm markers, while Jag1 inhibited mesoderm and endoderm inductions and displayed a dose-dependent contrasting effect on neural commitment (Fig. (Fig.2B).2B). For −LIF, however, Dll4 specifically promoted mesodermal commitment at both the doses and endodermal commitment at the low dose only (Fig. (Fig.2C).2C). In contrast, Jag1 augmented neural differentiation, inhibiting mesoderm (Fig. (Fig.2C).2C). Thus, Notch-induced differentiation in ESCs displayed an interesting ligand-biased lineage commitment and fate switching event between the mesoderm and the neuroectoderm.

FIG. 2.
Notch ligand bias and lineage commitment in ESCs. (A) qPCR showing an increase in Fgf5 expression with Jag1 (dose dependent) and decrease with Dll4 under the +LIF condition. For −LIF, only Dll4 displayed decreased expression compared to ...

Receptors and downstream genes involved during Notch-dependent commitment signaling.

The ligand-dependent fate decision in ESCs postulated a possible involvement of specific Notch receptors in individual commitment signals. Interestingly, expression of these receptors in ESCs was influenced by LIF status. ESCs expressed all four Notch receptors, with Notch1 and Notch4 showing relatively higher expressions under the +LIF condition (Fig. (Fig.3A).3A). However, with −LIF, both Notch1 and Notch4 showed comparatively low expression, while Notch2 and Notch3 increased (Fig. (Fig.3A).3A). Thus, involvement of Notch1 and Notch4 in maintenance and Notch3 in differentiation was speculated. These were further analyzed using the ligand-treated conditions to understand the receptors involved in a particular commitment process. The expression of Notch receptors remained quite variable in response to types and doses of ligands used and the LIF status (data not shown). Under the +LIF condition, Jag1 treatment at a lower dose (1 μg) promoted expression of almost all Notch receptors, while Dll4 (1 μg) increased Notch1 and decreased Notch4 compared to results for the control. However, at a higher dose, both the ligands decreased expression of most of the Notch receptors, except Notch4, whose expression was increased with Jag1 treatment. With −LIF, Jag1 decreased Notch2 and Notch3 expression, and Dll4, while decreasing both Notch1 and Notch2 expression, increased Notch4 expression at a higher dose (data not shown). Together, a specific expression profile of ligand-induced Notch receptors was indiscernible. However, this did not preclude the possible involvement of ligand-dependent receptor expression during Notch-dependent lineage commitment even in the presence of LIF. In fact, increased expression of Notch receptors in response to low Jag1 harmonized well with primitive ectoderm formation and increased Nes and decreased Bry expression, shown in Fig. Fig.2.2. Similarly, an increase in Notch4 expression under the −LIF condition seemed crucial in Dll4-induced inhibition of ectoderm and promotion of mesoderm commitments. Although linking the commitment signal with receptor expression seemed complex, the likely involvement of more than one receptor type was presumed to trigger the commitment signal in response to ligands.

FIG. 3.
qPCR analyses of Notch receptor and downstream gene expression. (A) Notch receptor expression profile in ESCs under the +LIF and −LIF conditions. Statistical significance was calculated, comparing the individual receptor expression under ...

Although receptor molecules differ, Notch signaling basically involves the Hes and Hey family of transcription corepressors downstream (15). Accordingly, the expression status of major downstream genes like Hes1, Hes2, Hes3, Hes5, and Hey1 was monitored. Hes2 and Hes3 expression in ESCs remained very low, with no difference even with ligand treatments (data not shown). However, there was a dose-dependent increase in Hes1 expression in response to ligands under the +LIF condition and with no appreciable difference under the −LIF condition compared to the respective controls (Fig. (Fig.3B).3B). While Jag1 increased Hes5 expression and decreased that of Hey1 compared with the control, Dll4 influenced the reverse pattern (Fig. (Fig.3B).3B). Interestingly, the Jag1-responsive Hes5 increase coincided with higher expression of Fgf5 and Nes (compare Fig. 2A to C and and3B),3B), thus hinting at neuroectodermal induction. Similarly, the Hey1 increase corresponded to higher Bry levels, suggesting mesodermal induction. Hence, Jag1-induced Notch signaling was deduced to promote neuroectodermal commitment in ESCs through Hes5, with Dll4 promoting mesoderm induction through Hey1 (Fig. 2A to C and and3B).3B). In fact, Hes5 has been reported to regulate NSC maintenance and inhibit neuronal differentiation (31). Although Hes1 expression increased with ligand treatments under the +LIF condition, it did not illustrate any ligand specificity. Hence, Hes1 involvement was postulated to accompany or accelerate specific factors involved in commitment processes rather than acting independently.

Notch promoted the generation of early neural progenitors.

Since Jag1-mediated Notch activation triggered neural commitment in ESCs during maintenance, in order to further substantiate this, we investigated its temporal influence during early neurogenic progression. In fact, the neurogenic developmental hierarchy involves neural commitment, progenitor specification, and generation of mature functional neural cells. Accordingly, further investigation was undertaken by following the established monoculture strategy for neural differentiation in ESCs (21) and with either Notch activation or inhibition using Jag1 or gSI, respectively, at different stages of differentiation. Our earlier study (21) demonstrated the Sox1+ population attaining a peak by day 4 and the Nes+ population during a 1-week time regimen (6 to 9 days), with both diminishing subsequently. While Sox1 is expressed in early neural progenitors (ENPs), Nes is expressed in all the neural progenitor populations. Hence, Notch involvement during early neurogenesis was ascertained by quantifying the Sox1+ and Nes+ populations at various stages during differentiation. An increase in the Sox1+ population on differentiation day 4 indicated a greater number of neural committed populations upon Notch activation (16.12% ± 1.18%) compared to the control (9.48% ± 1.28%) (Fig. (Fig.4A).4A). A similar trend was seen on day 7 (Fig. (Fig.4B).4B). In fact, the prolonged existence of ENPs on day 7 suggested Notch activation was helping in the maintenance of this population. However, Notch inhibition did not have any pronounced effect on the generation of Sox1+ ENPs compared to results for the vehicle control at both time points, though it decreased this population at a very early stage (day 2) of differentiation (Fig. (Fig.4C).4C). Monitoring time-dependent Notch inhibition during early differentiation determined that gSI repressed neural commitment during the initial time point (0 to 2 days), while no difference was observed (2 to 4 days) subsequently (Fig. (Fig.4D).4D). Interestingly, levels of Nes+ progenitors were higher upon gSI treatment (26.22% ± 3.00%) than upon Jag1 treatment (22.17% ± 1.9%) when monitored on day 7 (Fig. (Fig.4E),4E), with no difference seen on day 4 following treatments (see Fig. S2A in the supplemental material). Further, these populations were quantified on day 10 of differentiation, where neither activation nor inhibition had significant influence on Nes+ progenitors (gSI, 5.93% ± 0.52%; Jag1, 8.67% ± 0.90%) compared to the control (7.17% ± 1.04%) (see Fig. S2B), and Sox1 was undetected in cells by then. Further monitoring of the temporal influence of Jag1 demonstrated no variation in the Nestin population on the 1st week of differentiation (see Fig. S2C). Therefore, it was comprehended that Notch activation might not only promote ENPs but also help in their maintenance. However, Notch inhibition might promote overall neural progenitor populations in culture at 1 week, following their commitment in response to Notch activation. Collectively, this demonstrated an interesting paradigm of involvement of both Notch activation and inhibition in promoting generation of two different progenitor populations in a temporal and context-dependent manner.

FIG. 4.
Notch participation during neural progenitor generation. (A and B) Flow cytometry quantification of cells on the 4th (A) (d4) or 7th (B) (d7) day of differentiation revealed an increased population of Sox1+ cells upon Jag1 treatment compared to ...

To further substantiate Notch influence on neural progenitors, we used a nes-EGFP ESC clone (21) and quantified simultaneously both ENPs and late neural progenitors (LNPs) under Notch activation and inhibition conditions. These cells, upon differentiation, revealed populations with various EGFP intensities. Both ENPs and LNPs were discriminated by flow cytometry on the basis of their EGFP intensities. While ENPs expressed low EGFP (range, 101 to 102), LNPs displayed brighter fluorescence (range, >102) (21). The total EGFP population correlating with endogenous nestin expression (21) increased more significantly with gSI (44.89% ± 1.58%) than with Jag1 treatment (38.41% ± 5.43%) compared with their respective controls (Fig. (Fig.4F).4F). As far as the ENPs and LNPs were concerned, Jag1 increased the EGFPlow populations while gSI increased the EGFPhigh ones (Fig. (Fig.4F).4F). On histogram analysis, while a single intensity peak of the EGFPlow population was displayed with Jag1, a diffuse peak with more EGFPhigh populations was seen with gSI treatment (see Fig. S2D in the supplemental material). Thus, Notch resided at a critical juncture of neural development, where activation promoted commitment, proliferation, and maintenance of ENPs while inhibition paved the way for their further differentiation into LNPs. The knowledge gained would facilitate chalking out strategies for the enrichment of stage- and subtype-specific neural populations from ESCs in vitro.

Notch involvement during neural maturation.

Notch involvement in progenitors' generation was further extended to study its involvement during neural maturation. Accordingly, ESCs were differentiated and Notch signaling was inhibited at different time intervals during differentiation. Neural maturation was quantified on the basis of MAP2+ neurons and GFAP+ astrocyte populations during 14 to 16 days of differentiation. Notch inhibition at all the time regimens, except the initial one (0 to 2 days), promoted (~6 to 20%) the MAP2+ populations (Fig. (Fig.5A;5A; see also Fig. S3 in the supplemental material). Conversely, inhibition at all the time points diminished (~6 to 23%) the GFAP+ populations (Fig. (Fig.5B;5B; see also Fig. S3). Thus, inhibition at the initial time point attenuated the overall neural differentiation, corresponding with the decrease in the Sox1+ population as stated earlier. However, at other stages it specifically promoted neuronal differentiation at the expense of glial differentiation. In 1-week culture, too, neuronal differentiation was promoted, as seen by increased TuJ1+ (a marker for differentiating neurons) and MAP2+ populations following gSI treatment compared to the control (Fig. 5C and D; see also Fig. S4A). Notch inhibition also affected the morphology and structure of neurons. Neurons in the gSI treatment group displayed an increase in the length, branching, and number of neurites, demonstrating extensive neuronal arborization (7) (Fig. 5E and F). In fact, this act of neurite development was ameliorated when Notch was inhibited during the 2nd week. In parallel, neurogenesis was studied by activating Notch along these time intervals. As seen in Fig. Fig.5C,5C, Jag1 treatment decreased TuJ1+ cells in the 1st week (control, 25.75% ± 6.55%; Jag1, 6.92% ± 0.78%), while no GFAP+ cells were detected in both the control and treated groups on day 7. However, at the end of the 2nd week, the neural differentiation pattern with Jag1 treatment was comparable with that of the control (Fig. (Fig.6A;6A; see also Fig. S4B). Hence, together with the temporal profile of the Sox1+ and Nes+ population described, Notch activation delayed neural differentiation by specific promotion and maintenance of ENPs in a 1-week time regimen, while its inhibition expedited LNPs' differentiation into neurons and than glia and also promoted neuronal arborization in the 2nd week.

FIG. 5.
Notch involvement during neural maturation. (A) Quantification of the MAP2+ population on day 14 of differentiation revealed gSI treatment enhancing neuronal differentiation at all the time points except the initial one (0 to 2 days). (B) However, ...
FIG. 6.
Notch-mediated glial differentiation. (A) Quantification of MAP2+ and GFAP+ populations revealed the Jag1-CNTF combination was most suited for gliogenesis compared to other treatments, while neuronal differentiation was attenuated by the ...

Since Notch inhibition in our culture system attenuated glial differentiation and activation remained futile for the same, we presumed additional factors might be required for gliogenesis to materialize. Accordingly, ESCs subjected to Notch activation were exposed to the most common gliogenic inducers, such as ciliary neurotrophic factor (CNTF) and bone morphogenic protein 2 (BMP2), in the medium (10, 27). CNTF regulates glial differentiation of retinal stem cells (3), while BMPs promote astroglial differentiation of subventricular zone progenitor cells (10). ESCs in our study did not respond to individual treatments of BMP2 and CNTF toward neural differentiation. Hence, we investigated the influence of Notch in concert with either BMP2 or CNTF during neural differentiation. The strategy involved Notch activation during the 1st week, followed by CNTF or BMP2 treatment in the 2nd week. An extensive GFAP+ population was detected in the CNTF group, while BMP2 inhibited the overall neural differentiation (Fig. 6A and B). However, no difference in the GFAP+ population was observed when either Jag1 or CNTF treatment was given singly. When CNTF treatment was given subsequent to Notch inhibition, GFAP+ cells were attenuated, contrary to no appreciable difference being noticed in MAP2+ populations (Fig. (Fig.6B).6B). Thus, a concerted action between Notch and CNTF in promoting gliogenesis from ESCs was underscored. Collectively, Notch revealed its stage-specific influence on neural fate decision, progenitor formation, maturation, and neuron-glia judgment, suggesting its versatility and multifaceted function.


Notch-mediated differentiation signal in ESCs.

ESCs derived from the blastula-stage embryo are endowed with distinctive characteristics of self-renewal and pluripotency that allow these cells to serve as an ideal in vitro model for investigation of early embryonic development. We have investigated the role of Notch, a crucial cell fate modulator during development (6), deciphering its preferential ligand usage in dictating lineage specification from unspecified ESCs. The murine ESCs are normally maintained with LIF in serum-containing medium, where LIF inhibits nonneural commitment and BMP (present in serum) inhibits neural commitment (41). Though Notch follows a pathway independent of those of LIF and BMP, its interaction with these pathways has been found in various systems. In fact, Notch is shown to antagonize BMP, an active component during ESC maintenance (41), by inhibiting Id1 expression in endothelial cells (17). STAT3, the mediator of the LIF pathway, also interacts with Hes proteins (19) and is regulated by Notch signaling during NSC survival and differentiation (1). Our findings, illustrating the promotion of ESC differentiation by Notch activation despite the presence of LIF in the medium, suggest a critical role played by Notch signaling in the commitment process. Moreover, this also proposes the persistence of a possible dominating antagonistic effect of Notch signaling over maintenance signaling, thereby surmounting the latter's influence to promote differentiation in ESCs.

Ligand-dependent lineage commitment.

Ligand-mediated Notch activation in cells can be executed by culturing them either over the ligand-overexpressing fibroblasts (23) or over the immobilized ligands (39). In the present study, we followed the latter approach to avoid fibroblast contamination in culture. The differential effect of Notch ligands during lymphopoiesis in human has been reported earlier (18, 28). However, ours is the first report providing evidence of Notch ligand bias in lineage commitment and specification during early development, using ESCs as a model system. While Jag1 could mediate neural specification, Dll4 mediated mesodermal commitment in ESCs. In fact, we are pursuing further investigation to substantiate the Dll4-induced mesodermal specification in three-dimensional culture and the downstream events (unpublished data). This ligand-dependent lineage discrimination in ESCs suggests the plausible role played by Notch in the early commitment process by preferential ligand usage by a relatively homogenous population of cells. Undoubtedly, this can be exploited further to obtain lineage-specific differentiation from ESCs. Moreover, it would be interesting to explore the role played by other ligand molecules, which would provide functional significance underlying the development of diverse ligands and receptors in mammals during the course of evolution.

Intricacy in receptor expression during commitment.

Investigation of receptor involvement during this ligand-responsive lineage specificity demonstrated the participation of a particular group of receptors for a specific commitment process. This in turn suggested the existence of receptor groups that might vary with cell types, giving differential responses accordingly. Alagille syndrome is an autosomal disorder caused by mutations in the Jag1 gene (30), though the characteristic phenotypes cannot be mimicked in animal models by Jag1 mutation alone. Most of the phenotypes are acquired when both Jag1 and Notch2 are mutated (24). Similarly, during angiogenesis the formation of tip cells is regulated by Dll4-mediated Notch1 signaling (13). However, the tissue-specific distribution of Notch and the vascular phenotypes in mice deficient for them never coincide. Moreover, the targeted disruption of Dll1 and Notch2 in mice results in hemorrhage, though they are not detected in vessels (16). These reports illustrate the intricacies related to receptor expression and the ligand-mediated complex signaling mechanism involving an array of Notch receptors. Nonetheless, it culminates in the expression of specific downstream genes despite the intricacies. In the present study, Jag1-mediated Notch signaling resulted in enhanced Hes5 expression and diminished Hey1 expression, which led to the differentiation of ESCs toward a neural lineage. The reversal in the expression pattern of these genes was observed with Dll4 treatment, which mediated mesodermal commitment of ESCs. Thus, our study elucidated an interesting paradigm of specific ligand usage during Notch signaling, culminating in the regulation of specific downstream target genes and ultimately leading to differential commitment in ESCs.

Notch involvement during neural differentiation.

During development, Notch functions as a permissive signal rather than an instructive one (22). It maintains the NSC population both in vivo and in vitro (1). In fact, it was reported to be operational in NSCs, whereas intermediate neural progenitors showed attenuation in Notch signaling (26). Lowell et al. (23) have also demonstrated Notch involvement during neural commitment from ESCs. On the contrary, the presence of thin neuroepithelium in RBPJk null embryos despite their developmental impairment suggested the importance of Notch during later stages of development rather than early commitment (32). Our study illustrated the involvement of Notch activation in promoting ENPs in culture, while its inhibition promoted LNPs. Hence, Notch activation might be required at an early stage for neural commitment, subsequent to which Notch inhibition might led to overall neural progenitor generation and their further differentiation. Moreover, Notch inhibition did not completely deplete this population, suggesting a plausible balance between Notch activation and an inhibition requirement to exert developmental stage-specific influence. In fact, ENPs displayed the potential to form neurosphere-like structures (unpublished data). However, the nature of these cells remains elusive, since Notch promotes the generation of both NSCs (1, 26) and progenitors (22). Our investigation in fact revealed an intricate stage-specific temporal influence of Notch during early neurogenesis. While in 1-week culture Notch activation enhanced the neural progenitor population and delayed their subsequent differentiation, in 2-week cultures it did not exert a pronounced effect on neurogenesis. However, Notch inhibition at the early time point (0 to 2 days) impaired the overall neural differentiation, and this might be due to the reduction in Sox1+ ENPs in response to Notch inhibition, while the same at other time points promoted neuronal differentiation rather than differentiation into glia.

A number of recent reports also demonstrated gliogenic potential of Notch signaling in both the vertebrate and invertebrate systems (8, 9, 27). Transient Notch activation in neural crest stem cells inhibited their neurogenic potential and directed them to differentiate into glia (27). Our data for ESCs showed Notch activation promoting the generation of a GFAP+ population in culture only when it was followed by CNTF treatment, while neither alone could promote the same. Hence, it would be interesting to explore the molecular basis underlying this sequential contribution and the existing cross talk, if any. Collectively, Notch exerted a multifaceted influence in regulating neurogenesis by cell-cell interaction during various stages of differentiation from ESCs in vitro. This could be exploited further to generate the cells of interest, which would provide a great leap toward using ESCs in regenerative medicine. Incidentally, BMP2 treatment inhibited overall neural differentiation under similar conditions, though both CNTF and BMP2 were illustrated to promote gliogenesis (3, 10, 27). In fact, both these factors were reported to regulate gliogenesis through a STAT3-dependent signaling pathway in the central nervous system (4, 33). In ESCs, however, LIF activates STAT3 during their maintenance in the undifferentiated state, and LIF, in concert with BMP, modulates the same (41). Hence, Notch might be residing at a crossroad, guiding the ESCs maintained through STAT3 activation in the presence of LIF to a specific cell fate during differentiation depending on the specific Notch ligand usage.

Overall our study illustrated an interesting paradigm of Notch activation through ligand preference underlying the germ layer commitment and specification from unspecified ESCs in vitro. Contemplating a parallel between in vitro and in vivo scenarios, it may be comprehended that the organizational complexities may influence the interaction pattern between Notch ligands and receptors occurring in various permutations and combinations. A specific ligand-receptor combination may underlie a specific lineage commitment, although the signaling mechanism is evolutionarily conserved. The Notch ligand-biased lineage commitment from unspecified ESCs and the neural hierarchy with a stage-specific role played by Notch is represented in Fig. Fig.7.7. The knowledge gained should further pave the way for a better understanding of Notch influence during neural development and regeneration in vivo.

FIG. 7.
Schematic representation of stage-specific Notch involvement (↓, activation; [perpendicular], inhibition) during lineage commitment and neurogenic progression from ESCs.

Supplementary Material

[Supplemental material]


We thank Urban Lendahl (Karolinska Institute, Sweden) for the kind gift of the 12×CSL-luc reporter construct and Swapnil Walke and Pratibha Khot for helping with the FACS acquisition.

This work was partly supported by a project grant from the Department of Biotechnology (DBT), Government of India (BT/PR2993/MED/14/399/2002), and intramural funding from NCCS. S.K.R. is a graduate student supported by the Council of Scientific and Industrial Research (CSIR), India.


[down-pointing small open triangle]Published ahead of print on 12 February 2010.

Supplemental material for this article may be found at


1. Androutsellis-Theotokis, A., R. R. Leker, F. Soldner, D. J. Hoeppner, R. Ravin, S. W. Poser, M. A. Rueger, S. K. Bae, R. Kittappa, and R. D. McKay. 2006. Notch signalling regulates stem cell numbers in vitro and in vivo. Nature 442:823-826. [PubMed]
2. Artavanis-Tsakonas, S., M. D. Rand, and R. J. Lake. 1999. Notch signaling: cell fate control and signal integration in development. Science 284:770-776. [PubMed]
3. Bhattacharya, S., A. Das, K. Mallya, and I. Ahmad. 2008. CNTF-mediated signaling regulates neuronal versus glial differentiation of retinal stem cells/progenitors by concentration-dependent recruitment of MAPK and Jak-STAT pathways in conjunction with Notch signaling. Stem Cells 26:2611-2624. [PubMed]
4. Bonni, A., Y. Sun, M. Nadal-Vicens, A. Bhatt, D. Frank, I. Rozovsky, N. Stahl, G. Yancopoulos, and M. Greenberg. 1997. Regulation of gliogenesis in the central nervous system by the JAK-STAT signaling pathway. Science 278:477-483. [PubMed]
5. Chambers, I., D. Colby, M. Robertson, J. Nichols, S. Lee, S. Tweedie, and A. Smith. 2003. Functional expression cloning of Nanog, a pluripotency sustaining factor in embryonic stem cells. Cell 113:643-655. [PubMed]
6. Ehebauer, M., P. Hayward, and A. Arias. 2006. Notch, a universal arbiter of cell fate decisions. Science 314:1414-1415. [PubMed]
7. Franklin, J. L., B. E. Berechid, F. B. Cutting, A. Presente, C. B. Chambers, D. R. Foltz, A. Ferreira, and J. S. Nye. 1999. Autonomous and non-autonomous regulation of mammalian neurite development by Notch1 and Delta1. Cell 9:1448-1457. [PubMed]
8. Gaiano, N., and G. Fishell. 2002. The role of notch in promoting glial and neural stem cell fates. Annu. Rev. Neurosci. 25:471-490. [PubMed]
9. Grandbarbe, L., J. Bouissac, M. D. Rand, M. H. Angelis, S. Artavanis-Tsakonas, and E. Mohier. 2003. Delta-Notch signaling controls the generation of neurons/glia from neural stem cells in a stepwise process. Development 130:1391-1402. [PubMed]
10. Gross, R. E., M. F. Mehler, P. C. Mabie, Z. Zang, L. Sanchi, and J. A. Kessler. 1996. Bone morphogenetic proteins promote astroglial lineage commitment by mammalian subventricular zone progenitor cells. Neuron 17:595-606. [PubMed]
11. Guo, M., L. Y. Jan, and Y. N. Jan. 1996. Control of daughter cell fates during asymmetric division: interaction of Numb and Notch. Neuron 17:27-41. [PubMed]
12. Hébert, J. M., M. Boyle, and G. R. Martin. 1991. mRNA localization studies suggest that murine FGF-5 plays a role in gastrulation. Development 112:407-415. [PubMed]
13. Hellstrom, M., L. Phng, J. Hofmann, E. Wallgard, L. Coultas, P. Lindblom, J. Alva, A. Nilsson, L. Karlsson, N. Gaiano, K. Yoon, J. Rossant, M. Iruela-Arispe, M. Kalen, H. Gerhardt, and C. Betsholtz. 2007. Dll4 signalling through Notch1 regulates formation of tip cells during angiogenesis. Nature 445:776-780. [PubMed]
14. Hooper, C., J. P. Chapple, S. Lovestone, and R. Killick. 2007. The Notch-1 intracellular domain is found in sub-nuclear bodies in SH-SY5Y neuroblastomas and in primary cortical neurons. Neurosci. Lett. 415:135-139. [PMC free article] [PubMed]
15. Iso, T., L. Kedes, and Y. Hamamori. 2003. HES and HERP families: multiple effectors of the notch signaling pathway. J. Cell Physiol. 194:237-255. [PubMed]
16. Iso, T., Y. Hamamori, and L. Kedes. 2003. Notch signaling in vascular development. Arterioscler. Thromb. Vasc. Biol. 23:543-553. [PubMed]
17. Itoh, F., S. Itoh, M. J. Goumans, G. Valdimarsdottir, T. Iso, G. P. Dotto, Y. Hamamori, L. Kedes, M. Kato, and P. P. ten Dijke. 2004. Synergy and antagonism between Notch and BMP receptor signaling pathways in endothelial cells. EMBO J. 23:541-551. [PubMed]
18. Jaleco, A., H. Neves, E. Hooijberg, P. Gameiro, N. Clode, M. Haury, D. Henrique, and L. Parreira. 2001. Differential effects of Notch ligands Delta-1 and Jagged-1 in human lymphoid differentiation. J. Exp. Med. 194:991-1002. [PMC free article] [PubMed]
19. Kamakura, S., K. Oishi, T. Yoshimatsu, M. Nakafuku, N. Masuyama, and Y. Gotoh. 2004. Hes binding to STAT3 mediates crosstalk between Notch and JAK-STAT signalling. Nat. Cell Biol. 6:547-554. [PubMed]
20. Lenka, N., Z. J. Lu, P. Sasse, J. Hescheler, and B. K. Fleischmann. 2002. Quantitation and functional characterization of neural cells derived from ES cells using nestin enhancer-mediated targeting in vitro. J. Cell Sci. 115:1471-1485. [PubMed]
21. Lenka, N., and S. K. Ramasamy. 2007. Neural induction from ES cells portrays default commitment but instructive maturation. PLoS One 2:e1349. [PMC free article] [PubMed]
22. Louvi, A., and S. Artavanis-Tsakonas. 2006. Notch signalling in vertebrate neural development. Nat. Rev. Neurosci. 7:93-102. [PubMed]
23. Lowell, S., A. Benchoua, B. Heavey, and A. G. Smith. 2006. Notch promotes neural lineage entry by pluripotent embryonic stem cells. PLoS Biol. 4:e121. [PMC free article] [PubMed]
24. McCright, B., J. Lozier, and T. Gridley. 2002. A mouse model of Alagille syndrome: Notch2 as a genetic modifier of Jag1 haploinsufficiency. Development 129:1075-1082. [PubMed]
25. Mitsui, K., Y. Tokuzawa, H. Itoh, K. Segawa, M. Murakami, K. Takahashi, M. Maruyama, M. Maeda, and S. Yamanaka. 2003. The homeoprotein Nanog is required for maintenance of pluripotency in mouse epiblast and ES cells. Cell 113:631-642. [PubMed]
26. Mizutani, K., K. Yoon, L. Dang, A. Tokunaga, and N. Gaiano. 2007. Differential Notch signalling distinguishes neural stem cells from intermediate progenitors. Nature 449:351-355. [PubMed]
27. Morrison, S., S. Perez, Z. Qiao, J. Verdi, C. Hicks, G. Weinmaster, and D. Anderson. 2000. Transient Notch activation initiates an irreversible switch from neurogenesis to gliogenesis by neural crest stem cells. Cell 101:499-510. [PubMed]
28. Neves, H., F. Weerkamp, A. C. Gomes, B. A. Naber, P. Gameiro, J. D. Becker, P. Lucio, N. Clode, J. J. van Dongen, F. J. Staal, and L. Parreira. 2006. Effects of Delta1 and Jagged1 on early human hematopoiesis: correlation with expression of notch signaling-related genes in CD34+ cells. Stem Cells 24:1328-1337. [PubMed]
29. Nichols, J., B. Zevnik, K. Anastassiadis, H. Niwa, D. Klewe-Nebenius, I. Chambers, H. Schöler, and A. Smith. 1998. Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 95:379-391. [PubMed]
30. Oda, T., A. G. Elkahloun, B. L. Pike, K. Okajima, I. D. Krantz, A. Genin, D. A. Piccoli, P. S. Meltzer, N. B. Spinner, F. S. Collins, and S. C. Chandrasekharappa. 1997. Mutations in the human Jagged1 gene are responsible for Alagille syndrome. Nat. Genet. 16:242. [PubMed]
31. Ohtsuka, T., M. Ishibashi, G. Gradwohl, S. Nakanishi, F. Guillemot, and R. Kageyama. 1999. Hes1 and Hes5 as Notch effectors in mammalian neuronal differentiation. EMBO J. 18:2196-2207. [PubMed]
32. Oka, C., T. Nakano, A. Wakeham, J. L. de la Pompa, C. Mori, T. Sakai, S. Okazaki, M. Kawaichi, K. Shiota, T. W. Mak, and T. Honjo. 1995. Disruption of the mouse RBP-J kappa gene results in early embryonic death. Development 121:3291-3301. [PubMed]
33. Rajan, P., D. M. Panchision, L. F. Newell, and R. D. McKay. 2003. BMPs signal alternately through a SMAD or FRAP-STAT pathway to regulate fate choice in CNS stem cells. J. Cell Biol. 161:911-921. [PMC free article] [PubMed]
34. Reya, T., S. J. Morrison, M. F. Clarke, and I. L. Weissman. 2001. Stem cells, cancer, and cancer stem cells. Nature 414:105-111. [PubMed]
35. Schmitt, T. M., and J. Zúñiga-Pflücker. 2002. Induction of T cell development from hematopoietic progenitor cells by Delta-like-1 in vitro. Immunity 17:749-756. [PubMed]
36. Schroeter, E., J. Kisslinger, and R. Kopan. 1998. Notch-1 signalling requires ligand-induced proteolytic release of intracellular domain. Nature 393:382-386. [PubMed]
37. Tian, G., C. Sobotka-Briner, J. Zysk, X. Liu, C. Birr, M. Sylvester, P. Edwards, C. Scott, and B. Greenberg. 2002. Linear non-competitive inhibition of solubilized human γ-secretase by pepstatin A methylester, L685458, sulfonamides, and benzodiazepines. J. Biol. Chem. 277:31499-31505. [PubMed]
38. Tropepe, V., S. Hitoshi, C. Sirard, T. W. Mak, J. Rossant, and D. van der Kooy. 2001. Direct neural fate specification from embryonic stem cells. Neuron 30:65-78. [PubMed]
39. Varnum-Finney, B., L. Wu, M. Yu, C. Brashem-Stein, S. Staats, D. Flowers, J. Griffin, and I. Bernstein. 2000. Immobilization of Notch ligand, Delta-1, is required for induction of notch signaling. J. Cell Sci. 113:4313-4318. [PubMed]
40. Weinmaster, G., and C. Kintner. 2003. Modulation of notch signaling during somitogenesis. Annu. Rev. Cell Dev. Biol. 19:367-395. [PubMed]
41. Ying, Q. L., J. Nichols, I. Chambers, and A. Smith. 2003. BMP induction of Id proteins suppresses differentiation and sustains embryonic stem cell self-renewal in collaboration with STAT3. Cell 115:281-292. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)