SEARCH

SEARCH BY CITATION

Keywords:

  • Notch;
  • Presenilin;
  • Nuclear translocation;
  • Delta

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. CBF1-luciferase assay
  5. RESULTS
  6. The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation
  7. Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation
  8. CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2
  9. DISCUSSION
  10. Acknowledgements

It has been hypothesized that a presenilin 1 (PS1)-related enzymatic activity is responsible for proteolytic cleavage of the C-terminal intracellular protein of Notch1, in addition to its role in β-amyloid protein (Aβ) formation from the amyloid precursor protein (APP). We developed an assay to monitor ligand-induced Notch1 proteolysis and nuclear translocation in individual cells : Treatment of full-length Notch1-enhanced green fluorescent protein-transfected Chinese hamster ovary (CHO) cells with a soluble preclustered form of the physiologic ligand Delta leads to rapid accumulation of the C terminus of Notch1 in the nucleus and to transcriptional activation of a C-promoter binding factor 1 (CBF1) reporter construct. Nuclear translocation was blocked by cotransfection with Notch's physiologic inhibitor Numb. Using this assay, we now confirm and extend the observation that PS1 is involved in Notch1 nuclear translocation and signaling in mammalian cells. We demonstrate that the D257A and the D385A PS1 mutations, which had been shown previously to block APP γ-secretase activity, also prevent Notch1 cleavage and translocation to the nucleus but do not alter Notch1 trafficking to the cell surface. We also show that two APP γ-secretase inhibitors block Notch1 nuclear translocation with an IC50 similar to that reported for APP γ-secretase. Notch1 signaling, assessed by measuring the activity of CBF1, a downstream transcription factor, was impaired but not abolished by the PS1 aspartate mutations or γ-secretase inhibitors. Our results support the hypotheses that (a) PS1-dependent APP γ-secretase-like enzymatic activity is critical for both APP and Notch processing and (b) the Notch1 signaling pathway remains partially activated even when Notch1 proteolytic processing and nuclear translocation are markedly inhibited. The latter is an important finding from the perspective of therapeutic treatment of Alzheimer's disease by targeting γ-secretase processing of APP to reduce Aβ production.

The majority of early-onset autosomal dominant familial Alzheimer's disease (AD) is associated with mutations in presenilin (PS) genes (PS1 and PS2). PS mutations result in markedly elevated levels of β-amyloid protein (Aβ) deposition in the brains of AD patients (Lemere et al., 1996 ; Gomez-Isla et al., 1999). Aβ, a 40-42-amino acid peptide, is generated from the transmembrane portion of a large single-pass transmembrane protein, amyloid precursor protein (APP), by action of β-secretase and γ-secretase. Recently PS1 has been implicated as being intimately associated with γ-secretase activity. In mice lacking PS1 the level of Aβ is decreased (De Strooper et al., 1998) owing to decreased activity of γ-secretase. APP γ-secretase activity can also be diminished by mutations in two transmembrane aspartate residues in PS1 (D257A and D385A) or by use of peptidomimetic γ-secretase inhibitors (Wolfe et al., 1999) and was completely abolished by mutating aspartate residues in both PS1 and PS2 (Kimberly et al., 2000).

Notch1 is a cell surface receptor important in cell fate decisions during development (Artavanis-Tsakonas et al., 1995). Interaction of Notch with its ligand on adjacent cells activates downstream transcription factors of the CSL family [C-promoter binding factor 1 (CBF1), Su(H), and Lag1] (Furukawa et al., 1991 ; Schweisguth and Posakony, 1992 ; Artavanis-Tsakonas et al., 1995, 1999 ; Schweisguth, 1995 ; Lu and Lux, 1996). The signaling is believed to be initiated when Notch, a large type I integral membrane protein with a single transmembrane domain, is cleaved intracellularly near the cell surface, and the proteolytically released intracellular domain, termed Notch1 C-terminal intracellular domain, (NICD), translocates to the nucleus (Goodbourn, 1995 ; Kopan et al., 1996 ; Weinmaster, 1997 ; Schroeter et al., 1998).

A relationship between Notch1 and PS1 was initially suggested by the presence of a genetic interaction between Sel-12 and Lin-12, the homologues of PS1 and Notch in Caenorhabditis elegans (Levitan and Greenwald, 1995, 1998). PS1 null mice have a phenotype similar to that of Notch1 null mice (Shen et al., 1997 ; Wong et al., 1997), as do PS1-deficient Drosophila (Struhl and Greenwald, 1999 ; Ye et al., 1999). This interaction has been further supported by the observations that Notch1 and PS1 are coexpressed during development in the CNS in the mouse (Berezovska et al., 1997) and colocalize in embryos of fruitfly (Ye and Fortini, 1998) and in mammalian neurons (Berezovska et al., 1998). PS1 and Notch have been coimmunoprecipitated from mammalian and Drosophila cells (Ray et al., 1999b). Finally, PS1 has been shown to modulate Notch1's effect on neurite outgrowth in postmitotic mammalian neurons (Berezovska et al., 1999a).

Recent biochemical data show that, in PS1 null cells, constitutively active forms of Notch1 in which the extracellular ligand binding domain was deleted fail to undergo proteolytic cleavage and nuclear translocation, suggesting a direct role for PS1 in Notch proteolytic cleavage and release from the membrane (De Strooper et al., 1999 ; Song et al., 1999). These data give rise to the hypothesis that a PS1-dependent enzymatic activity, or PS1 itself, is responsible for processing both APP and Notch1 (De Strooper et al., 1998, 1999 ; Song et al., 1999). Alternatively, other reports suggest a role for PS1 in events upstream of proteolytic processing, such as protein trafficking or folding (Naruse et al., 1998 ; Guo et al., 1999 ; Katayama et al., 1999 ; Ye et al., 1999).

To distinguish among these possibilities, we developed an assay for direct visualization of rapid Notch1 nuclear translocation after ligand-dependent activation of Notch1 using a full-length Notch1-enhanced green fluorescent protein (EGFP) fusion protein activated by a soluble form of the Notch1 ligand, Delta. On ligand treatment, the C terminus of Notch1 rapidly translocates to the nucleus and induces the transcriptional activation of a CBF1-dependent reporter gene. Notch1 nuclear translocation can be prevented by cotransfection with Numb, a physiological inhibitor of Notch1 (Frise et al., 1996 ; Guo et al., 1996 ; Spana and Doe, 1996). Moreover, we confirm and extend the previous observation that PS1 is critical for Notch1 processing, now using full-length Notch1 activated by its physiological ligand, and also report additional evidence that implicates PS1/γ-secretase in Notch1 proteolytic processing in a manner analogous to APP processing. We show that overexpression of the D257A or D385A PS1 mutations (which diminish APP γ-secretase activity) prevents generation and nuclear translocation of NICD. It is important that neither the D257A nor the D385A PS1 mutations affected trafficking and insertion of full-length Notch1 into the plasma membrane. In addition, two peptidomimetic inhibitors of the γ-secretase cleavage of APP also inhibit Notch1's ligand-induced nuclear translocation in a dose-dependent manner, arguing in favor of a PS1 effect on Notch1 processing rather than trafficking. Finally, a functional assay using a CBF1-luciferase reporter construct suggests relative preservation of Notch1 signaling under conditions where the generation and nuclear translocation of free NICD are markedly inhibited.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. CBF1-luciferase assay
  5. RESULTS
  6. The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation
  7. Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation
  8. CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2
  9. DISCUSSION
  10. Acknowledgements

Plasmids

Full-length Notch1, N1(FL), cloned into pEF-Bos vector was prepared as described (Shawber et al., 1996). To generate the N1(FL)-EGFP fusion construct, tagged with EGFP on the Notch1 C terminus, full-length Notch1 cDNA was digested out of pEF-Bos vector and cloned into an EGFP vector (pEGFPN3 ; Clontech, Palo Alto, CA, U.S.A.). Similarly, two truncated constitutively active Notch1 constructs, containing the membrane spanning region and the cytoplasmic portion (N1▵EC ; 5,374-7,836 bp) or just the cytoplasmic signaling portion of Notch1 (NICD ; 5,476-7,836 bp), were cloned into the pEGFP-N3 vector. N1(FL) tagged with hemagglutinin (HA) on its extracellular domain (Rand et al., 2000) was cloned into pBABE (Morgenstern and Land, 1990). The m-Numb plasmid (Zhong et al., 1996) was a generous gift from Dr. Y. N. Jan (Howard Hughes Medical Institute, University of California, San Francisco).

CBF1-luciferase assay

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. CBF1-luciferase assay
  5. RESULTS
  6. The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation
  7. Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation
  8. CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2
  9. DISCUSSION
  10. Acknowledgements

We cotransfected cells with Notch1 (FL or ▵EC), a CBF1-firefly luciferase reporter (Hsieh et al., 1996), and an internal control Renilla luciferase vector pRL-tk (Dual-Luciferase Reporter Assay System ; Promega, Madison, WI, U.S.A.). Luciferase activity was measured after 24 h. Transfection with an empty vector served as a negative control for the experiment.

Cell culture and transient transfection

Chinese hamster ovary (CHO) cells (ATCC, Manassas, VA, U.S.A.) and stably transfected CHO cells expressing either wild-type (WT) human PS1 or one of the transmembrane Asp[RIGHTWARDS ARROW]Ala PS1 mutations (D257A or D385A) (Wolfe et al., 1999) were transiently transfected with 1 μg of N1(FL)-EGFP cDNA/ml of culture medium using SuperFect reagents (Qiagen, Valencia, CA, U.S.A.). Conditioned medium (CM) was isolated from transfected human kidney fibroblast 293T cells secreting a soluble Fc-conjugated Delta (Dl-Fc), which has previously been shown to activate Notch signaling (Wang et al., 1998) ; CM from wild-type 293T cells was used as a negative control. To activate secreted Dl-Fc, we preclustered it by incubating the supernatant containing Dl-Fc with an anti-Fc antibody (Jackson IRL, West Grove, PA, U.S.A.) for 1 h at 4°C. At 24 h after the transfection, activated Delta-Fc (Dl-Fc) was added to the cultures (1 part of CM with or without Dl-Fc to 2 parts of culture medium) for different intervals (0, 1, 3, 5, 15, 30, and 60 min and 6-24 h). For Numb and Notch1 cotransfection experiments, we used a mixture of 1 μg/ml Numb cDNA and 1 μg/ml N1(FL) cDNA to transfect WT PS1 CHO cells.

Immunocytochemistry and confocal microscopy

We noted that EGFP fluorescence could be readily observed by fluorescence microscopy in the nucleus of N1(FL)-EGFP-transfected cells treated with D1-Fc for 14-24 h but that at earlier time points the amount of fluorescence in the nucleus was too weak to detect visually. We therefore amplified the EGFP signal by anti-green fluorescent protein immunocyto-chemistry and increased detection by using quantitative confocal microscopy.

Cells incubated with preclustered D1-Fc were fixed in 4% paraformaldehyde, permeabilized with 0.5% Triton X-100 (for anti-EGFP and anti-Notch but not for anti-HA immunostaining), blocked with normal goat serum, and immunostained with an anti-EGFP (Clontech), anti-Notch1 (Aster et al., 1997), or anti-HA antibodies (Babco), followed by incubation with Cy3-labeled secondary antibody. The images of all transfected cells within a visual field immunostained with anti-green fluorescent protein antibody were collected using constant settings on a confocal microscope (Bio-Rad model 1024). Each image was a result of accumulation of six laser scans of the cell to increase the signal detection. The intensity of immunofluorescence (photon counts) in the nucleus of the transfected cell was analyzed using Adobe Photoshop software. The data were collected from four experiments. On average, 100-150 immunopositive cells for each time point were quantified.

To assess the effect of D257A and D385A PS1 mutations on Notch trafficking, we used an N1(FL) construct tagged with HA on the N-terminal (extracellular) side, which has previously been shown to be biologically active (Rand et al., 2000). After transfection, cells were immunostained for HA with or without membrane permeabilization with 0.5% Triton X-100.

Western blot analysis

Western blot analysis was performed as described by Berezovska et al. (1998) using anti-Notch1 antibody (Aster et al., 1997). Enhanced chemiluminescence was used to detect immunoreactivity. Bands on films were quantitated using a Bio-Rad Multi-Analyst System.

γ-Secretase inhibitor treatment

Two difluoroketone peptidomimetics, MW167 and CM115 [referred to as compound 1 and compound 11, respectively, by Wolfe et al. 1998, 1999)], were designed on the basis of the APP structure to inhibit selectively γ-secretase activity. The γ-secretase inhibitors were diluted in dimethyl sulfoxide (DMSO) and added to the cells 3 h after transfection for an additional 24 h. MW167 or CM115 (1, 10, 50, and 100 μM) was used for dose-response curve (see Fig. 5e) ; corresponding amounts of DMSO (maximum of 1% DMSO for 100 μM dose) were added to sister cultures as negative controls.

image

Figure 5. γ-Secretase inhibitors abolish ligand-induced Notch1 nuclear translocation. N1(FL)-EGFP-transfected WT PS1 CHO cells were incubated with (a) “empty” 293T supernatant or 293T supernatant containing (b) preclustered DI-Fc, or DI-Fc and either (c) CM115 or (d) MW167 APP γ-secretase inhibitors for 24 h at 20 μM (Wolfe et al., 1998, 1999). Bar = 50 μm. e : Inhibition of the ligand-induced accumulation of Notch1-EGFP in the nucleus was dose-dependent with an IC50 of ~20-30 μM. The cells were incubated for 24 h with 0.01% DMSO vehicle alone or with 1, 10, 50, and 100 μM CM115 or MW167 in 0.01% DMSO. Data are mean ± SD (bars) values. *p < 0.01, **p < 0.001 by ANOVA test compared with DMSO. f : Inhibition of NICD generation in N1▵EC-transfected WT PS1 CHO cells treated with 20 μM CM115. Top panel : Immunoblot. Bottom panel : Ratio of NICD to unprocessed N1▵EC. Data are mean ± SD (bars) values.

Download figure to PowerPoint

The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. CBF1-luciferase assay
  5. RESULTS
  6. The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation
  7. Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation
  8. CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2
  9. DISCUSSION
  10. Acknowledgements

To characterize full-length Notch1/Delta interactions and to evaluate translocation of NICD to the nucleus, we transfected CHO cells with N1(FL)-EGFP. At 24 h after transfection, the cells expressed N1(FL)-EGFP in the cell body and cell surface. No signal was detectable in the nucleus. We then added preclustered Dl-Fc, or similarly prepared fractions of control media, for various intervals. At 24 h after incubation with preclustered Dl-Fc we observed robust accumulation of green fluorescence in the nucleus of ~30-50% of all transfected cells (Fig. 1b) but in none of the cells exposed to the control media (Fig. 1a). This pattern of intense nuclear fluorescence is identical to the pattern observed after transfection with N1▵EC-EGFP (Fig. 1c) and NICD-EGFP (Fig. 1d), which are constitutively active, ligand-independent forms of Notch1, but differs from the diffuse cytoplasmic/nuclear staining pattern seen after transfection with a plasmid containing EGFP alone.

image

Figure 1. Transfection of Notch1-EGFP into CHO cells. a : CHO cells transfected with full-length Notch1 and treated for 24 h with control 293T media. There was no difference between untreated cultures and cultures treated with control media. b : N1(FL)-EGFP-transfected cells treated for 24 h with CM containing preclustered Notch1 ligand, Dl-Fc. c and d : Cells transfected with N1▵EC-EGFP and NICD-EGFP, respectively. Arrows indicate green fluorescence in the nucleus. Bar = 50 μm. e : Time course of Notch1 nuclear translocation induced by preclustered Dl-Fc in CHO cells. The intensity of EGFP-immunoreactivity in the nucleus of CHO cells treated with CM containing preclustered Dl-Fc (starting at 3-5 min to 24 h) was significantly greater than EGFP immunoreactivity in the nucleus of the cells incubated with “empty” 293T CM. The data represent photon counts (intensity of the EGFP immunoreactivity) in the nucleus measured on a 0-255 scale of optical intensity and are mean ± SD (bars) values. *p <0.05, **p<0.001 by ANOVA.

Download figure to PowerPoint

Absence of the signal in the nucleus at earlier time points after exposure to preclustered Dl-Fc could indicate that the amount of C-terminal fragment proteolytically released from N1(FL)-EGFP is below the level of detection for EGFP with standard fluorescence microscopy. Therefore, we developed a more sensitive technique by amplifying the EGFP signal with immunocytochemical staining with an anti-green fluorescent protein antibody and assaying the amount of nuclear fluorescence using quantitative confocal microscopy. Using the enhanced detection method, we were able to detect fluorescence in the nucleus as early as 3-5 min after adding preclustered Dl-Fc. The intensity of immunofluorescence in the nucleus increased rapidly over the first hour and reached saturation in this assay (optical density reading of 255) by 4-6 h ; the intensity in the nucleus remained saturated (using the high sensitivity settings selected for the early time points) through 24 h (Fig. 1e). Treatment of cells with 293T cell CM not containing Dl-Fc did not significantly increase EGFP immunoreactivity in the nucleus even using these highly sensitive detection techniques.

Several approaches were used to confirm that the nuclear green fluorescence is due to translocation of NICD fused to EGFP. First, immunostaining with C-terminal Notch1 antibody revealed increased Notch1 immunoreactivity in the nucleus of ~30-50% of N1(FL)-EGFP-transfected cells treated with a ligand and in a majority of N1▵EC-EGFP-transfected cells, a percentage of cells similar to that observed by EGFP immunofluorescence (data not shown). Second, we contransfected Nl(FL) with Numb, and upstream modulator of the Notch signaling pathway that is known to bind to the C terminus of Notch1 and prevent activation of Notch1 signal transduction (Guo et al., 1996 ; Wakamatsu et al., 1999). Numb significantly decreased the amount of ligand-induced nuclear translocation of Notch1 by 70 ± 10% (average ± SD photon counts in the nuclei of 30 cells, p < 0.01) even after 24 h of incubation with preclustered Dl-Fc (Fig. 2). Similar results were observed in three independent experiments. In addition, we demonstrated that ligand treatment led to transcriptional activation of a CBF1 reporter construct, indicating activation of downstream components of the Notch signaling pathway. Measuring CBFl-luciferase activity in cells transfected with N1(FL) and treated with Dl-Fc or in cells transfected with constitutively active truncated N1▵EC showed four to six and eight to 12 times activation, respectively, in comparison with empty vector-transfected cells (see Fig. 6).

image

Figure 2. Contransfection of Numb with N1(FL) suppresses nuclear translocation of Notch1-EGFP. Confocal microscopy shows N1(FL)-EGFP (a) before adding its ligand, preclustered Dl-Fc, or (b) 24 h after adding preclustered Dl-Fc, (c) cotransfection of N1(FL)-EGFP and Numb, and (d) cotransfection of N1(FL)-EGFP and Numb after 24 h of treatment with preclustered Dl-Fc. Bar = 50 μm.

Download figure to PowerPoint

image

Figure 6. Transcriptional activation of a CBF1-luciferase promoter in CHO cells transfected with N1(FL) or N1▵EC. There is relative preservation of Notch1 signaling 24 h after transfection with constitutively active, membrane-tethered Notch1 in cells overexpressing PS1 aspartate mutations or in WT PS1 cells treated with γ-secretase inhibitors. a : The WT PS1, D257A, or D385A PS1 cells were transfected with an empty vector, full-length Motch1 and treated with Dl-Fc [N1(FL)+Dl], or truncated Notch1 (N1▵EC). b : WT PS1 cells were treated for 24 h with 0.01% DMSO or 75 μM MW167 or 75 μM CM115 γ-secretase inhibitor in 0.01% DMSO. c : CHO cells overexpressing WT PS1 and PS2 or aspartate mutant PS1 and PS2 (D257A PS1 and D366A ; 2A2) were transiently transfected with N1▵EC. To measure activation of CBF1, the cells were contransfected with Notch1 and CBF1-luciferase (firefly luciferase) together with pRL-tk reporter vector (Renilla luciferase). The ratio of firefly luciferase activity to Renilla luciferase activity normalizes CBF1 luciferase activity to the efficiency of transfection in every cell line/experiment. The mean ± SD (bars) luciferase activation in Notch-transfected cells relative to that in empty vector-transfected cells was calculated and presented as a fold activation. The number in each column indicates the number of experiments. *p < 0.05, **p < 0.01 by ANOVA with post hoc Fisher's PLSD test.

Download figure to PowerPoint

Notch1 nuclear translocation is impaired by D257A or D385A PS1 mutations

Overexpression of PS1 mutated in either of two transmembrane aspartate residues (D257A and D385A) led to replacement of native PS1 and decreased activity as loss of function mutations markedly diminishing APP γ-secretase activity (Wolfe et al., 1999). Here we tested the hypothesis that the same PS1 asparate mutations would interfere with Notch1 processing and signaling. CHO cells stably expressing WT PS1, D257A PS1, or D385A PS1 were transiently transfected with N1(FL)-EGFP and then incubated with preclustered Dl-Fc. There was no significant difference in the amount or subcellular localization of Notch1-EGFP expression in WT PS1 compared with mutant PS1 (D257A or D385A)-expressing cells (Fig. 3a-c) ; all three lines show N1-EGFP signal in the cytoplasm and on the plasma membrane, with abundant accumulation of green fluorescence endoplasmic reticulum and Golgi apparatus but none in the nucleus.

image

Figure 3. Activation of Notch1(FL) with its ligand, Dl-Fc, in CHO cells overexpressing WT PS1, D257A PS1, or D385A PS1 mutations. a-f : Montage of confocal micrographs of living N1(FL)-EGFP-transfected CHO cells incubated for 24 h with 293T supernatant with or without preclustered Dl-Fc. a-c : WT PS1, D257A, and D385A CHO cells, respectively, incubated with “empty” 293T supernatant. d-f : WT PS1, D257A, and D385A CHO cells, respectively, incubated with preclustered Dl-Fc. Arrows in (d) indicate accumulation of green fluorescence (EGFP-tagged C terminus of Notch1) in the nuclei. Inset in (d) represents images of living cells 14 h after Dl-Fc treatment representing “intermediate” stages of accumulation of EGFP-tagged Notch1 C terminus in the nucleus. Arrowheads in (e) and (f) indicate absence of green fluorescence in the nuclei. Bar = 50 μm. g : Time course of Notch1 nuclear translocation in CHO cells overexpressing WT PS1 (□) or D257A PS1 (○) and D385A PS1 (▵) mutations. Similar to D257A and D385A PS1-overexpressing cells, there was no increase in fluorescence in the nucleus in WT PS1 CHO cells after incubation with control 293T media (⋄). Data are mean ± SD (bars) values. *p < 0.01, **p < 0.001 by ANOVA. h : D257A and D385A PS1 mutations inhibit proteolytic release of NICD. Top panel : The indicated cell lines were transfected with N1▵EC, lysed 24 h later, and immunoblotted with a Notch1 C-terminal antibody. The proteolytic generation of NICD was observed in WT PS1-expressing cells but was barely observed in D257A and D385A PS1-expressing cells. An equal amount of protein was loaded on each lane. The arrowhead indicates a nonspecific band at 36 kDa acting as an internal loading control. Bottom panel : Ratio of NICD to total N1▵EC in different cell lines (quantitative densitometry of NICD and N1▵EC bands on western blot), standardized to the WT PS1 cell ratio. Data are mean ± SD (bars) values. **p < 0.001 by ANOVA.

Download figure to PowerPoint

Starting 1 min after adding preclustered Dl-Fc to WT PS1 CHO cells, fluorescent signal could be detected in the nucleus of WT PS1 cells by quantitative anti-EGFP immunocytochemistry and digital confocal microscopy (Fig. 3). Addition of control 293T CM did not induce any translocation of fluorescent signal to the nucleus, even 24 h after addition.

In contrast to the results obtained in WT PS1-overexpressing cells, overexpression of either D257A or D385A PS1 completely blocked nuclear accumulation of Notch1 during 60 min of incubation of the cells with Dl-Fc, as assessed by quantitative anti-EGFP immunocytochemistry (Fig. 3). Moreover, in contrast to robust nuclear signal in the WT PS1 cells, no specific fluorescent signal was detected in the nucleus of D257A or D385A cells even 24 h after exposure of the cells to the ligand (Fig. 3d-f), indicating that both aspartate mutations in PS1 prevent, and not simply delay, ligandinduced Notch1-EGFP nuclear translocation/accumulation.

To test further the hypothesis that PS1 aspartate mutations inhibit the intramembranous cleavage of Notch, we analyzed the proteolytic release of NICD in WT PS1-, D257A PS1-, or D385A PS1-expressing cells transiently transfected with the constitutively active membrane spanning Notch1 construct, N1ΔEC. We observed the generation of NICD from N1ΔEC in WT PS1-expressing cells (Fig. 3h). The amount of cleaved NICD was significantly reduced in cells expressing either one of the aspartate mutant PS1, in comparison with that in WT PS1 cells. The ratio of NICD to unprocessed N1ΔEC was reduced by 77 and 70% in D257A and D385A PS1 cells, respectively, in comparison with that in cells expressing WT PS1 (Fig. 3h). The substantial diminution of proteolytic processing in the aspartate mutant cell lines may explain the absence of visible nuclear accumulation of Notch1-EGFP in these cells.

Notch1 cellular localization is not altered by D257A or D385A PS1 mutations

An alternative explanation for the lack of nuclear translocation of Notch1 in response to ligand in the D257A and D385A PS1 cells would be if Notch1 did not reach the cell surface. Similar arguments about PS1 effects on protein trafficking have been raised in other systems (Naruse et al., 1998 ; Guo et al., 1999). To test the hypothesis that the D257A and D385A PS1 mutations impact Notch's trafficking to the cell surface, we used an N1(FL) construct bearing an extracellular domain HA tag (Rand et al., 2000). Transfection of WT PS1, D257A PS1, and D385A PS1 CHO cells was followed by immunostaining for HA. Similar staining of the extracellular HA, assessed by immunostaining without addition of permeabilizing agents, was observed in WT PS1, D257A PS1, and D385A PS1 CHO cells (Fig. 4a-c). Permeabilization of the cells with Triton X-100 before immunostaining also showed no difference in the pattern of staining or levels of expression among the three cell lines (Fig. 4d-f). In addition, western blot analysis of WT PS1, D257A, and D385A PS1 cells transfected with N1(FL) showed the normal furin-like cleavage to form mature Notch1 heterodimer with the 110-kDa band in addition to 250-kDa full-length Notch1 in all cell lines tested, in accord with the observations in Drosophila PS1 null cells of Guo et al. (1999). These data suggest that the D257A and D385A mutations do not alter Notch1 receptor maturation or surface expression.

image

Figure 4. D257A and D385A PS mutations do not interfere with Notch1 receptor trafficking to the cell surface and its insertion into plasma membrane. HA immunoreactivity was assayed in CHO cells without (a-c) and with (d-f) Triton X-100 treatment in WT PS1 (a and d), D257A PS1 (b and e), and D385A PS1 (c and f) cells. Bar = 50 μm.

Download figure to PowerPoint

Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. CBF1-luciferase assay
  5. RESULTS
  6. The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation
  7. Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation
  8. CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2
  9. DISCUSSION
  10. Acknowledgements

Another approach to characterizing the proteolytic activity that occurs after Notch1 is activated by its ligand is to use pharmacologic inhibitors. We used two well-characterized peptidomimetic APP γ-secretase inhibitors (Wolfe et al., 1998, 1999) to test the hypothesis that APP γ-secretase inhibition and Notch1 proteolysis had similar pharmacological profiles. Ligand-induced Notch1 nuclear translocation was markedly diminished, in a dose-dependent manner, when N1(EL)-transfected, WT PS1-overexpressing cells were treated with either of the APP γ-secretase inhibitors but not with vehicle (DMSO) alone (Fig. 5a). A dose-response curve (Fig. 5e) shows the IC50 for each to be 20-30 μM. These results were confirmed by western blot analyses, which revealed that the generation of NICD was markedly impaired in WT CHO cells treated with 20 μM CM115 (Fig. 5f).

CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. CBF1-luciferase assay
  5. RESULTS
  6. The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation
  7. Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation
  8. CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2
  9. DISCUSSION
  10. Acknowledgements

To study whether the Notch signaling pathway could still be activated when only very small amounts of NICD could be detected by western blot and in the absence of detectable Notch1 nuclear translocation, we measured transcriptional activation of a CBF1-luciferase reporter construct by N1(FL)+D1 or by N1▵EC in CHO cells. We found that CBF1-luciferase activity was significantly higher, as expected, in N1▵EC-transfected cells in comparison with N1(FL)-transfected cells treated with D1-Fc (Berezovska et al., 1999b). In addition, CBF1-luciferase activity in D257A PS1- or D385A PS1-overexpressing cells was reduced by ~70 and 35%, respectively, compared with control WT PS1-overexpressing cells (Fig. 6a).

The remaining CBF1 activity could be the result of very small amounts of NICD generated by residual PS1 activity, PS2 activity, or alternative proteases. However, treatment with the APP γ-secretase inhibitor CM115 or MW167 at 75 μM, which also reduced Notch1 nuclear translocation to undetectable levels, reduced CBF1-luciferase activity only by 48 and 35%, respectively (Fig. 6b). Moreover, even cell lines stably overexpressing both mutant PS1 (D257A) and PS2 (D366A) [2A2 (Kimberly et al., 2000)] show a marked reduction but not complete abolishment of CBF1-luciferase activity (Fig. 6c).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. CBF1-luciferase assay
  5. RESULTS
  6. The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation
  7. Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation
  8. CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2
  9. DISCUSSION
  10. Acknowledgements

We have developed a method to assess Notch1 activation at the cell surface at the single cell level. Incubation of N1(FL)-EGFP-transfected CHO cells with a preclustered soluble form of Delta, a physiological ligand of Notch, leads to detectable nuclear translocation of the C-terminal portion of Notch1 within minutes and to trans-activation of a CBF1 reporter construct. Cotransfection with Numb, a known physiologic inhibitor of Notch activation (Frise et al., 1996 ; Guo et al., 1996 ; Spana and Doe, 1996 ; Wakamatsu et al., 1999), prevents Notch1 Nuclear translocation. The pattern of nuclear translocation of fluorescence observed after treatment of Notch1-EGFP-transfected cells with D1-Fc is seen with both EGFP and Notch immunostaining and is essentially identical to the pattern of distinct nuclear localization observed after transfection of the cells with the ligand-independent, constitutively active constructs NICD-EGFP and N1▵EC-EGFP. Enhanced immunoreactivity for EGFP and quantitative confocal microscopy were sensitive enough to detect nuclear signal after just several minutes of ligand treatment. From these data we conclude that N1(FL)-EGFP is expressed at the cell surface and that the C-terminal domain can be cleaved and translocated to the nucleus in a ligand-dependent fashion.

We used this assay, as well as using constitutively active N1▵EC, to examine the role of PS1 in Notch1 processing and signaling. We tested the hypothesis that ligand-activated Notch1 proteolysis was carried out by an enzymatic activity similar or identical to APP γ-secretase. Two lines of evidence are presented. It has been shown previously that specific genetic manipulations of PS1 (D257A or D385A PS1 mutations) block APP γ-secretase (Wolfe et al., 1999). We demonstrate that overexpressing either of these PS1 mutations also abolishes the ligand-dependent nuclear translocation of Notch1 to undetectable levels in a morphological assay. These appear to be loss-of-function mutations, which, because of the “replacement phenomenon” in which endogenous PS1 is down-regulated when PS1 constructs are overexpressed, are analogous to dominant negative mutations with respect to APP proteolysis (Wolfe et al., 1999) ; our data suggest that they also act in this fashion with regard to Notch C-terminal proteolysis and nuclear translocation.

An alternative hypothesis is that the D257A or D385A PS1 mutants interfere with trafficking of membrane proteins (Levitan and Greenwald, 1998 ; Naruse et al., 1998). Autophosphorylation of the TrkB receptor after BDNF treatment is reduced by 90% in PS1 null neurons, possibly because of alterations in maturation of the receptor (Naruse et al., 1998). In the absence of PS1 in Drosophila, processing of Notch in the Golgi apparatus by furin is normal, but a redistribution of the cellular location with increased Notch content on the cell surface has been reported (Guo et al., 1999). Prevention of ligand-induced proteolysis and nuclear translocation could occur if there was an alteration in Notch1 reaching the cell surface in the D257A or D385A PS1 cells. We therefore examined Notch1 in the D257A or D385A PS1 mutant cells and found it to be present on the cell surface, analogous to nontransfected or WT PS1 cells. These data suggest that the D257A or D385A PS1 mutations do not affect Notch1 access to the plasma membrane and support the idea that these mutations impact Notch1 signaling by interfering with Notch1 proteolysis and/or nuclear translocation. Similar results were obtained using pulsechase and biotinylation assays to show that PS1 binds to Notch1 in the endoplasmic reticulum/Golgi apparatus and then cotransports to the plasma membrane as a complex ; aspartate mutations did not affect the ability of PS1 either to bind Notch1 or to traffic to the plasma membrane (Ray et al., 1999a). Whether the interaction between PS1 and either Notch1 or APP that leads to proteolysis occurs at the cell surface or in an internalized compartment (Annaert et al., 1999 ; Katayama et al., 1999) is not certain, but our data support the idea that the effects of the D257A and D385A PS1 mutations are downstream to Notch's ligand binding at the cell surface.

In addition to these experiments in which we introduced specific PS1 mutants, we also present pharmacological data testing the hypothesis that APP γ-secretaselike activity is important for ligand-induced Notch1 cleavage. We demonstrate that two peptidomimetic compounds based on APP structure and previously shown to inhibit APP γ-secretase (Wolfe et al., 1999) also effectively inhibit Notch1 cleavage/nuclear translocation with identical IC50 values. This result is consistent with the data of De Strooper et al. (1999) showing inhibition of cleavage of constitutively active truncated Notch (mNotch▵EC) by the APP γ-secretase inhibitor MW167.

In accord with the results from the morphological nuclear translocation assay, we found that generation of the signaling C-terminal fragment of Notch1 was markedly diminished by overexpression of D257A or D385A PS1 mutations or by the APP γ-secretase inhibitors. However, in contrast to these morphological or biochemical assays, we found surprising preservation of Notch1 signaling as assayed with a CBF1-luciferase reporter construct. This suggests at least two possibilities : (a) that even small amounts of NICD, near the limits of detection, are sufficient to produce a substantial CBF1 signal, or (b) that generation of a Notch fragment and translocation to the nucleus are not obligatory steps in Notch signaling. The fact that there is an apparent correlation between the amount of NICD generated in WT PS1, D257A PS1, and D385A PS1 cell lines and the degree of CBF1 activation in these cell lines (Figs. 3h and 6a) argues in favor of the first suggestion. This is in agreement with data presented by Schroeter et al. (1998) showing that reduction in the amount of NICD produced parallels reduction in HES-1 promoter activation. An additional issue is that the difference between the morphological/biochemical assays and the functional assay may reflect the relative sensïtivities and ranges of the assays. Consideration of the marked embryonic lethal phenotype of PS1 null animals, which is analogous to Notch1 null animals (Shen et al., 1997 ; Wong et al., 1997), shows that residual non-PS1 Notch1 signaling is not sufficient for normal development. Nonetheless, the relative preservation of Notch1 signaling even in the setting of substantial γ-secretase inhibition suggests that partial inhibition of γ-secretase function in a clinical setting may not adversely effect Notch1 signaling. Our data show that using γ-secretase inhibitors in a concentration that practically abolishes Aβ production (Wolfe et al., 1998) reduces but does not block Notch1 signaling in CHO cells in the CBF1-luciferase assay.

Despite the relative preservation of Notch1 signaling seen in the CBF1-luciferase assay compared with morphological assay, it is clear that alteration of both PS1 and PS2 by introduction of aspartate mutations and pharmacological APP γ-secretase inhibitors do significantly reduce both proteolytic cleavage of the C-terminal domain and CBF1-luciferase activity, showing that Notch-PS interactions are important components in Notch1 processing and signaling. This is in agreement with results from the HES-1 assay in PS1-deficient fibroblast cells (Song et al., 1999) but contrasts with the data reported by Berechid et al. (1999) that PS1 deficiency had no measurable effect on Notch1 signaling in PS1 null fibroblasts. The latter researchers postulated that the preserved HES-1 activation might be due to partial redundancy of PS genes, with PS2 remaining active in PS1-deficient fibroblasts. Our data using CHO cells expressing both D257A PS1 and D366A PS2 double mutants support this idea.

Taken together, the experiments using PS1 mutations and pharmacological data strongly support several conclusions : (a) Ligand-induced cleavage of full-length Notch1 occurs via a PS1-dependent γ-secretase enzymatic activity identical to that having APP γ-secretase activity. (b) D257 and D385 in PS1 are each independently critical for this activity. (c) Aspartate mutations in PS1 do not affect trafficking of Notch1 to the cell surface. (d) Generation and nuclear translocation of the NICD are nearly abolished by aspartate mutations in PS1 and APP γ-secretase inhibitors in both morphological and biochemical assays. (e) Notch1 signaling, as assessed by a CBF1-luciferase functional assay, is significantly reduced but not abolished under the same conditions.

We hypothesize that the principal-function of the PSs in development is to interact with Notch1 and related integral membrane proteins and that its processing of APP gradually leads to the accumulation of the proteolytic product, Aβ, in the brain in postreproductive life. Compounds such as those studied here could be used to inhibit the generation of Aβ from APP to treat or prevent AD. Our current data suggest that they only partially inhibit Notch1 signaling ; whether this would contribute to adverse treatment effects remains to be determined.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. CBF1-luciferase assay
  5. RESULTS
  6. The C-terminal portion of Notch1 rapidly translocates to the nucleus after ligand binding and induces CBF1 transcriptional activation
  7. Peptidomimetic APP γ-secretase inhibitors also block Notch C-terminal cleavage and nuclear translocation
  8. CBF1 activation is relatively preserved in CHO cells treated with γ-secretase inhibitors and in cells expressing aspartate mutant PS1 and PS2
  9. DISCUSSION
  10. Acknowledgements

We thank Dr. Y. N. Jan (University of California, San Francisco) for m-Numb plasmids, Chad Moore for providing CM115 protease inhibitor, and C. J. Smith for help with cell cultures. This work was supported by grants PO AG 15379 and AG 14744 from the National Institutes of Health.

  • 1
    Annaert W., Levesque L., Craessaerts K., Dierinck I., Snellings G., Westaway D., George-Hyslop P., Cordell B., Fraser P., De Strooper B. (1999) Presenilin 1 controls gamma-secretase processing of amyloid precursor protein in pre-Golgi compartments of hippocampal neurons.J. Cell Biol. 147,277294.
  • 2
    Artavanis-Tsakonas S., Matsuno K., Fortini M.E. (1995) Notch signaling.Science 268,225232.
  • 3
    Artavanis-Tsakonas S., Rand M.D., Lake R.J. (1999) Notch signaling : cell fate control and signal integration in development.Science 284,770776.
  • 4
    Aster J., Robertson E., Hasserjian R., Turner J., Kieff E., Sklar J. (1997) Oncogenic forms of Notch1 lacking either the primary binding site for RBP-Jκ or nuclear localization signal sequences retain the ability to associate with RBP-Jκ and activate transcription.J. Biol. Chem. 272,1133611343.
  • 5
    Berechid B.E., Thinakaran G., Wong P.C., Sisodia S.S., Nye J.S. (1999) Lack of requirement for presenilin1 in Notch1 signaling.Curr. Biol. 9,14931496.
  • 6
    Berezovska O., Xia M.Q., Page K., Wasco W., Tanzi R.E., Hyman B.T. (1997) Developmental regulation of presenilin mRNA expression parallels Notch expression.J. Neuropathol. Exp. Neurol. 56,4044.
  • 7
    Berezovska O., Xia M., Hyman B. (1998) Notch is expressed in adult brain, is coexpressed with presenilin-1, and is altered in Alzheimer disease.J. Neuropathol. Exp. Neurol. 57,738745.
  • 8
    Berezovska O., Frosch M., McLean P., Knowles R., Koo E., Kang D., Shen J., Lu F.M., Lux S.E., Tonegawa S., Hyman B.T. (1999a) The Alzheimer-related gene presenilin 1 facilitates Notch1 in primary mammalian neurons.Mol. Brain Res. 69,273280.
  • 9
    Berezovska O., McLean P., Knowles R., Frosh M., Lu F.M., Lux S.E., Hyman B.T. (1999b) Notch1 inhibits neurite outgrowth in postmitotic primary neurons.Neuroscience 93,433439.
  • 10
    De Strooper B., Saftig P., Craessaerts K., Vanderstichele H., Guhde G., Annaert W., Von Figura K., Van Leuven F. (1998) Deficiency of presenilin-1 inhibits the normal cleavage of amyloid precursor protein.Nature 391,387390.
  • 11
    De Strooper B., Annaert W., Cupers P., Saftig P., Craessaerts K., Mumm J.S., Schroeter E.H., Schrijvers V., Wolfe M.S., Ray W.J., Goate A., Kopan R. (1999) A presenilin-1-dependent γ-secretase-like protease mediates release of Notch intracellular domain. Nature 398,518521.
  • 12
    Frise E., Knoblich J.A., Younger-Shepherd S., Jan L.Y., Jan Y.N. (1996) The Drosophila Numb protein inhibits signaling of the Notch receptor during cell-cell interaction in sensory organ lineage.Proc. Natl. Acad. Sci. USA 93,1192511932.
  • 13
    Furukawa T., Kawaichi M., Matsunami N., Ryo H., Nishida Y., Honjo T. (1991) The Drosophila RBP-J kappa gene encodes the binding protein for the immunoglobulin J kappa recombination signal sequence.J. Biol. Chem. 266,2333423340.
  • 14
    Gomez-Isla T., Growdon W., McNamara M., Nochlin D., Bird T., Arango J., Lopera F., Kosik K.S., Lantos P.L., Cairns N.J., Hyman B.T. (1999) The impact of different presenilin 1 and presenilin 2 mutations on amyloid deposition, neurofibrillary changes and neuronal loss in the familial Alzheimer's disease brain : evidence for other phenotype-modifying factors.Brain 122,17091719.
  • 15
    Goodbourn S. (1995) Notch takes a short cut.Nature 377,288289.
  • 16
    Guo M., Jan L.Y., Jan Y.N. (1996) Control of daughter cell fates during asymmetric division : interaction of Numb and Notch.Neuron 17,2741.
  • 17
    Guo Y., Livne-Bar I., Zhou L., Boulianne G. (1999) Drosophila presenilin is required for neuronal differentiation and affects Notch subcellular localization and signaling.J. Neurosci. 19,84358442.
  • 18
    Hsieh J.J., Henkel T., Salmon P., Robey E., Peterson M.G., Hayward S.D. (1996) Truncated mammalian Notch1 activates CBF1/RBPJ κ-repressed genes by a mechanism resembling that of Epstein-Barr virus EBNA2. Mol. Cell. Biol. 16,952959.
  • 19
    Katayama T., Imaizumi K., Sato N., Miyoshi K., Kudo T., Hitomi J., Morihara T., Yoneda T., Gomi F., Mori Y., Nakano Y., Takeda J., Tsuda T., Itoyama Y., Murayama O., Takashima A., George-Hyslop P., Takeda M., Tohyama M. (1999) Presenilin-1 mutations downregulate the signalling pathway of the unfoldedprotein response.Nat. Cell Biol. 1,479485.
  • 20
    Kimberly W.T., Xia W., Rahmati T., Wolfe M.S., Selkoe D.J. (2000) The transmembrane aspartates in presenilin 1 and 2 are obligatory for γ-secretase activity and amyloid β-protein generation. J. Biol. Chem. 275,31733178.
  • 21
    Kopan R., Schroeter E., Weintraub H., Nye J. (1996) Signal transduction by activated mNotch : importance of proteolytic processing and its regulation by the extracellular domain.Proc. Natl. Acad. Sci. USA 93,16831688.
  • 22
    Lemere C.A., Lopera F., Kosik K.S., Lendon C.L., Ossa J., Saido T.C., Yamaguchi H., Ruiz A., Martinez A., Madrigal L., Hincapie L., Arango J.C., Anthony D.C., Koo E.H., Goate A.M., Selkoe D.J., Arango J.C. (1996) The E280A presenilin 1 Alzheimer mutation produces increased A beta 42 deposition and severe cerebellar pathology.Nat. Med. 2,11461150.
  • 23
    Levitan D. & Greenwald I. (1995) Facilitation of lin-12-mediated signalling by sel-12, a Caenorhabditis elegans S182 Alzheimer's disease gene.Nature 377,351355.
  • 24
    Levitan D. & Greenwald I. (1998) Effect of Sel-12 presenilin on Lin-12 localization and function inCaenorhabditis elegans. Development 125,35993606.
  • 25
    Lu F.M. & Lux S.E. (1996) Constitutively active human Notch1 binds to the transcription factor CBF1 and stimulates transcription through a promoter containing a CBF-responsive element.Proc. Natl. Acad. Sci. USA 93,56635667.
  • 26
    Morgenstern J.P. & Land H. (1990) Advanced mammalian gene transfer : high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line.Nucleic Acids Res. 18,35873596.
  • 27
    Naruse S., Thinakaran G., Luo J., Kusiak J., Tomita T., Iwatsubo T., Qian X., Ginty D.D., Price D.L., Borchelt D.R., Wong P.C., Sisodia S.S. (1998) Effect of PS1 deficiency on membrane protein trafficking in neurons.Neuron 21,12131221.
  • 28
    Rand M., Grimm L., Artavanis-Tsakonas S., Patriub V., Blacklow S., Sklar J., Aster J.C. (2000) Calcium depletion dissociates and activates heterodimeric Notch receptors.Mol. Cell. Biol. 20,18251835.
  • 29
    Ray W., Yao M., Mumm J., Schroeter E., Saftig P., Wolfe M., Selkoe D.J., Kopan R., Goate A.M. (1999a) Cell surface presenilin-1 participates in the gamma-secretase-like proteolysis of Notch.J. Biol. Chem. 274,3680136807.
  • 30
    Ray W., Yao M., Mowotny P., Mumm J., Zhang W., Wu J., Kopan R., Goate A.M. (1999b) Evidence for a physical interaction between presenilin and Notch.Proc. Natl. Acad. Sci. USA 96,32633268.
  • 31
    Schroeter E., Kisslinger J., Kopan R. (1998) Notch-1 signalling requires ligand-induced proteolytic release of intracellular domain.Nature 393,382386.
  • 32
    Schweisguth F. (1995) Suppressor of Hairless is required for signal reception during lateral inhibition in the Drosophila pupal notum.Development 121,18751884.
  • 33
    Schweisguth F. & Posakony J. (1992) Suppressor of Hairless, the Drosophila homolog of the mouse recombination signal-binding protein gene, controls sensory organ cell fates.Cell 69,11991212.
  • 34
    Shawber C., Nofziger D., Hsieh J.J., Lindsell C., Bogler O., Hayward D., Weinmaster G. (1996) Notch signaling inhibits muscle cell differentiation through a CBF1 -independent pathway.Development 122,37653773.
  • 35
    Shen J., Bronson R.T., Chen D.F., Xia W., Selkoe D.J., Tonegawa S. (1997) Skeletal and CNS defects in presenilin-1-deficient mice.Cell 89,629639.
  • 36
    Song W., Nadeau P., Yang X., Shen J., Yankner B. (1999) Proteolytic release and nuclear translocation of Notch-1 are induced by presenilin-1 and impaired by pathogenic presenilin-1 mutations.Proc. Natl. Acad. Sci. USA 96,69596963.
  • 37
    Spana E.P. & Doe C.Q. (1996) Numb antagonizes Notch signaling to specify sibling neuron cell fates.Neuron 17,2126.
  • 38
    Struhl G. & Greenwald I. (1999) Presenilin is required for activity and nuclear access of Notch inDrosophila. Nature 398,522525.
  • 39
    Wakamatsu Y., Maynard T., Jones S., Weston J. (1999) NUMB localizes in the basal cortex of mitotic avian neuroepithelial cells and modulates neuronal differentiation by binding to NOTCH-1.Neuron 23,7181.
  • 40
    Wang S., Sdrulla A., Sibio G., Bush G., Nofziger D., Hicks C., Weinmaster G., Barres B.A. (1998) Notch receptor activation inhibits oligodendrocyte differentiation.Neuron 21,6375.
  • 41
    Weinmaster G. (1997) The ins and outs of notch signaling.Mol. Cell. Neurosci. 9,91102.
  • 42
    Wolfe M., Citron M., Diehl T., Xia W., Donkor I., Selkoe D. (1998) A substrate-based difluoro ketone selectively inhibits Alzheimer's gamma-secretase activity.J. Med. Chem. 41,69.
  • 43
    Wolfe M., Xia W., Ostaszewski B.L., Thekla S.D., Kimberly W.T., Selkoe D.J. (1999) Two transmembrane aspartates in presenilin-1 required for presenilin endoproteolysis and γ-secretase activity. Nature 398,513517.
  • 44
    Wong P.C., Zheng H., Chen H., Becher M.W., Sirinathsinghji D.J., Trumbauer M.E., Chen H.Y., Price D.L., Van der Ploeg L.H., Sisodia S.S. (1997) Presenilin 1 is required for Notch1 and DII1 expression in the paraxial mesoderm.Nature 387,288292.
  • 45
    Ye Y. & Fortini M. (1998) Characterization of Drosophila presenilin and its colocalization with Notch during development.Mech. Dev. 79,199211.
  • 46
    Ye Y., Lukinova N., Fortini M.E. (1999) Neurogenic phenotypes and altered Notch processing in Drosophila presenilin mutants.Nature 398,525529.
  • 47
    Zhong W., Feder J., Jiang M., Jan L., Jan Y. (1996) Asymmetric localization of a mammalian Numb homolog during mouse cortical neurogenesis.Neuron 17,4353.