JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M502769200 on March 29, 2005

J. Biol. Chem., Vol. 280, Issue 23, 22462-22472, June 10, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/23/22462    most recent
M502769200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, H.
Right arrow Articles by Kim, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, H.
Right arrow Articles by Kim, K.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Presenilin-1 D257A and D385A Mutants Fail to Cleave Notch in Their Endoproteolyzed Forms, but Only Presenilin-1 D385A Mutant Can Restore Its {gamma}-Secretase Activity with the Compensatory Overexpression of Normal C-terminal Fragment*

Hangun Kim{ddagger}, Hyunkyoung Ki{ddagger}, Hee-Sae Park§, and Kwonseop Kim{ddagger}

From the {ddagger}College of Pharmacy and the §School of Biological Sciences and Technology, Chonnam National University, Gwangju 500-757, Korea

Received for publication, March 14, 2005


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The enzyme {gamma}-secretase is involved in the cleavage of several type I membrane proteins, such as Notch 1 and amyloid precursor protein. Presenilin-1 (PS-1) is one of the critical integral membrane protein components of the {gamma}-secretase complex and is processed endoproteolytically, generating N- and C-terminal fragments (NTF and CTF, respectively). PS-1 is also known to incorporate into a high molecular weight complex by binding to other {gamma}-secretase components such as Nicastrin, Aph-1, and Pen-2. Mutations on PS-1 can alter the effects of {gamma}-secretase on its many substrates to different extents. Here, we showed that PS-1 mutants have a different activity for Notch cleavage, which depended on the PS-1 mutation site. We demonstrated that defective PS-1 mutants located in CTF, i.e. D385A and C410Y, could restore their {gamma}-secretase activities with the compensatory overexpression of wild type CTF or of minimal deleted CTF (amino acids 349–467). However, the defective PS-1 D257A mutant could not restore their {gamma}-secretase activities with the compensatory overexpression of wild type NTF. In comparison, both D257A NTF and D385A CTF could abolish the {gamma}-secretase activity of wild type and pathogenic PS-1 mutants. We also showed that PS-1 NTF but not CTF forms strong high molecular weight aggregates in SDS-PAGE. Taken together, results have shown that NTF and CTF integrate differently into high molecular weight aggregates and that PS-1 Asp-257 and Asp-385 have different accessibilities in their unendoproteolyzed conformation.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Alzheimer disease (AD)1 is the most common cause of dementia, producing an impairment of memory and cognitive abilities which is gradual in onset but relentless in progression. AD is observed in more than 10% of individuals over the age of 65, but its exact pathogenic mechanism is elusive. AD is associated with massive accumulation of fibrillary aggregates in various cortical and subcortical regions of the brain, which include intracellular neurofibrillary tangles and extracellular amyloid plaques (1, 2). Familial Alzheimer disease (FAD) accounts for 5–10% of AD cases, and ~50% of these have been definitely linked to mutations in the three genes that encode the amyloid precursor protein, presenilin 1 (PS-1), and presenilin 2 (PS-2). To date, more than 150 different FAD-linked PS-1 mutations have been identified and account for the vast majority of FAD-related mutations (2–5).

PS-1 is composed of 467 amino acids and is located predominantly in the membranes of the rough endoplasmic reticulum and, to a lesser extent, in the early Golgi complex (3, 6, 7). PS-1 is processed endoproteolytically to generate 27–28-kDa N- and 18–20-kDa C-terminal fragments, which are tightly regulated by competition of limiting cellular factors and might be a rate-limiting step in PS maturation (2). The cleavage of PS-1 occurs preferentially at amino acid positions 291 and 298 encoded by exon 9, a region where several PS-1 missense and deletion mutations are clustered (2, 8). PS-1 is known to play a pivotal role in the catalytic activity of {gamma}-secretase, a membrane-associated enzyme complex consisting of PS, Nicastrin, Aph-1, and Pen-2 (9–11). Studies with artificial and spontaneously occurring PS-1 mutants reveal that the activity of {gamma}-secretase is critically dependent on the presence of two intact aspartate residues, Asp-257 and Asp-385, and on the endoproteolysis of PS-1. However, it is still unclear whether the defective PS-1 mutants, D257A and D385A, result from the changes in the catalytic domain or from the lack of their own endoproteolysis (12–18).

The components of the {gamma}-secretase complex act upon each other: PS-1 is required for maturation of Nicastrin (19) and affects the stability of Pen-2 by inhibiting the proteosome-mediated degradation (20, 21). In turn, Pen-2 coordinately regulates proteolytic processing of PS-1 with Aph-1 (22) and is required for stabilization of the PS NTF/CTF heterodimer within the {gamma}-secretase complex (23). Aph-1 interacts with PS and Nicastrin and is required for intramembrane proteolysis of several {gamma}-secretase substrates (24, 25).

A number of type I transmembrane proteins have been identified as {gamma}-secretase substrates, including amyloid precursor protein (26, 27), CD44 (28, 29), E-cadherin (30), ErbB4 (31), and Notch (32–34). Among these, Notch is the receptor for Jagged and Delta, which are critically required for a variety of signaling events both during embryogenesis and in adulthood (35). PSs are involved in the proteolytic processing (S3 cleavage) of Notch, which results in the release of its intracellular domain (NICD), which in turn acts as a transcription factor in the nucleus after associating with either a member of the CSL family of DNA-binding proteins or with LEF-1 (36, 37). The Notch ligands, Delta1 and Jagged, are sequentially cleaved by an ADAM protease and {gamma}-secretase, indicating that {gamma}-secretase plays a pivotal role in the Delta1-Notch signaling pathway (38, 39). In addition to its role in neuronal cell differentiation, Notch 1 is known to compete with amyloid precursor protein for {gamma}-secretase and down-regulates PS-1 gene expression (40). However, the exact role of Notch 1 in the progression of AD remains to be demonstrated.

More than 150 FAD-linked PS-1 mutations have been identified to date and are suggested to affect {gamma}-secretase activity to different extents according to substrate levels. In this report, we used four different mouse embryonic fibroblast (MEF) cell lines with different PS-1 and -2 levels, as well as several PS-1 mutants to investigate the effect of exogenous PS-1 on the cleavage of Notch. Because the presence of endogenous PS-1 or -2 was sufficient to cleave Notch, PS-1/2-/- MEFs were used. We demonstrated that the cleavage of Notch depends on the mutation site in PS-1. With the use of NTF and CTF fragments, we clearly showed that the defects in PS-1 D257A and D385A are caused by changes in their catalytic domains rather than by the lack of endoproteolysis. The defects in PS-1 D385A mutants could be restored by the compensatory overexpression of its normal counterpart, wild type (wt) CTF. In contrast to D385A, defects in PS-1 D257A could not be restored by wt NTF. The minimal domain of wt CTF required to maintain {gamma}-secretase activity is located between residues 349 and 467. Our results indicate that NTF and CTF showed different HMW aggregate compositions and displayed different restoring effects on PS-1 D257A and D385A, as well as on different PS-1 mutants, suggesting that their regulation or function can be different.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plasmids—The wt and mutant PS-1s (Y115H, M146V, M146/D257A, D257A, L286V, D385A, C410Y), PS-1 fragments (NTF: 1–298, 1–131, 1–200, 1–260, 131–298; CTF: 299–467, 299–349, 299–418, 349–418, 349–467, 418–487 amino acids), and PS-1 D257A mutated NTF and PS-1 D385A mutated CTF and its fragments (299–349, 299–418, 349–418, 349–468, 418–487) in pEGFP-C1 were generated by PCR amplification and cloned into pEGFP-C1 vector (Clontech). The wt PS-1 in pEGFP-N3 was generated by PCR amplification and cloned into pEGFP-N3 vector (Clontech). After construction, the sequences were confirmed by DNA sequencing and found to be in-frame. The Notch constitutive active mutant was a kind gift from Dr. Raphael Kopan (Washington University, St. Louis). The Nicastrin construct was generated by PCR and cloned into pcDNA3-V5 (Invitrogen) vector. A HA-tagged hPen-2 in pcDNA3.1D/V5-His-TOPO (Invitrogen) was a kind gift from Dr. Takeshi Iwatsubo (University of Tokyo).

Antibodies—Peptide antibody (Ab) specific for green fluorescent protein (GFP 1:100) was purchased from Clontech, BD Biosciences. Polyclonal Ab (anti-cleaved Notch 1 (Val-1744) or anti-NICD) recognizing cleaved Notch 1 (Val-1744) only when cleaved between Gly-1743 and Val-1744, but not full-length Notch 1 or Notch 1 cleaved at other positions, was purchased from Cell Signaling Technology, Inc. The HA epitope was detected with media from 12CA5 hybridoma. The V5 epitope was detected with anti-V5 Ab (Invitrogen). The FLAG epitope was detected with anti-FLAG M2 Ab (Sigma). For the quantification of protein loading, anti-tubulin Ab was used (Sigma). For the detection of the C-terminal fragment (CTF) of PS-1, Presenilin 1 (C-20) was employed (Santa Cruz Biotechnology, Inc.).

Cell Culture and Transfection—MEF cell lines, PS-1-/- (deficient in PS-1), PS-2-/- (deficient in PS-2), PS-1/2-/- (deficient in both PS-1 and PS-2), and PS-1/2+/+ (expressing both PS-1 and PS-2), were obtained from Dr. De Strooper and cultured in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37 °C with 5% CO2. Cells were transfected with the use of Lipofectamine Plus reagent (Invitrogen) according to the manufacturer's instructions.

Immunoprecipitations—Immunoprecipitations were performed on lysates from transfected cells. Cells were lysed in lysis buffer containing 10 mM Tris-Cl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.2 mM sodium orthovanadate, 0.2 mM phenylmethysulfonyl fluoride, and protease inhibitor mixture. Lysates were incubated with primary antibodies for 16 h at 4 °C and pulled down with protein G-Sepharose (Amersham Biosciences). Immunoprecipitated proteins were eluted at 95 °C for 2 min with 50 µl of 2 sample buffer (0.1 M Tris-HCl, pH 6.8, 0.2 M dithiothreitol, 4% SDS, 20% glycerol, 0.2% bromphenol blue, 1.43 M {beta}-mercaptoethanol) and analyzed by immunoblotting as described below.

Immunoblotting—Cells were harvested with radioimmune precipitation assay lysis buffer (1% Triton X-100, 0.5% deoxycholate, 0.1% SDS, 150 mM NaCl, 50 mM Tris-Cl, 2 mM EDTA, 1 mM sodium orthovanadate, 10 µg/ml leupeptin, 5 µg/ml aprotinin, 1 mM phenylmethysulfonyl fluoride, protease inhibitor mixture). Cell lysates were quantified using the BCA assay kit (Pierce) and electrophoresed using 8% Tris-glycine gels (Novex). Proteins were transferred to a polyvinylidene difluoride membrane (Amersham Biosciences) that was subsequently probed with antibodies. Bound antibodies were visualized with horseradish peroxidase-conjugated secondary antibodies followed by visualization with enhanced chemiluminescence (ECL) Western blotting detection reagents (Amersham Biosciences). Even loadings on each lane were confirmed using direct blue 71 staining kits (EZBiopaQ, Co., Ltd.). For the NICD/PS-1 and NICD/tubulin ratio, a quantification of the protein bands was performed with a HP scanjet 3570c scanner (Hewlett-Packard) interfaced to an IBM computer, and image analysis was performed using the TINA 2.09 software program (Raytest).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Notch Cleavage Depends on the Presence of Endogenous PS-1 and PS-2 and on the Type of PS-1 Mutants—Small amounts of endogenous PS-1 and PS-2 can interfere with the estimation of exogenously expressed PS-1. We used four different cell lines to investigate the effects of endogenous PS-1 and PS-2 on the cleavage of Notch. In the PS-1/2+/+ MEFs that express both endogenous PS-1 and PS-2 and in the PS-1-/- and PS-2-/- MEFs that express only endogenous PS-2 and PS-1, respectively, Notch cleavage occurred regardless of the presence of exogenous PS-1 (Fig. 1A). Furthermore, NICD, the cleaved form of Notch, was still generated in PS-1/2+/+ cell lines even after the overexpression of PS-1 D385A, the catalytically inactive form of PS-1 (data not shown). However, in PS-1/2-/- MEFs having no endogenous PS-1 and PS-2, Notch cleavage occurred only when exogenous PS-1 was overexpressed (Fig. 1A). Therefore, we concluded that PS-1/2-/- MEF cell lines having no endogenous PSs should be employed to investigate the effect of exogenously overexpressed PS-1.

Using PS-1/2-/- cell lines and various mutant PS-1 constructs tagged with GFP, we observed different Notch cleavage patterns (top panel, Fig. 1B). The overexpressed wild types and mutant PS-1s were localized mainly in the endoplasmic reticulum and Golgi but were also found throughout the cytoplasm (data not shown). We analyzed the expression levels of different PS variants (middle panel, Fig. 1B), which were quantified and related to the amounts of NICD formed (bottom panel, Fig. 1B). Each of four spontaneous PS-1 mutants cleaved Notch to a different extent. As shown in Fig. 1B, PS-1 with a point mutation at M146V cleaved Notch slightly more than wt PS-1 did, whereas PS-1 with a mutation at Y115H and L286V cleaved Notch to a lesser extent than did wt PS-1. Interestingly, the spontaneous PS-1 C410Y mutant had defects in Notch cleavage. When the PS-1 putative catalytic site residues, Asp-257 and Asp-385, were artificially point mutated to Ala, the resulting PS-1 mutants could not cleave Notch. Likewise, M146V/D257A double mutated PS-1 did not cleave Notch, indicating that the presence of both Asp-257 and Asp-385 is required for Notch cleavage. When this set of experiments was performed in PS-1/2+/+ cells, all PS-1 mutants that were defective in cleaving Notch in PS-1/2-/- MEFs generated NICD regardless of the PS-1 mutant type (data not shown), supporting the notion that PS-1/2-/- MEFs should be used to investigate the effect of exogenous PS-1 constructs.



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 1.
The cleavage of Notch depends on the presence of endogenous PSs and on the type of PS-1 mutation. A, the effects of endogenously existing PS-1 and PS-2 on Notch cleavage were investigated using anti-cleaved Notch 1 Ab. B, Notch cleavage patterns brought about by different types of PS-1 mutants were investigated in the PS-1/2-/- cell lines with anti-NICD Ab (top panel). The expression levels of the different PS variants were analyzed with anti-PS-1 Ab (middle panel), quantified using the TINA 2.09 software program (Raytest), and related to the amounts of NICD formed. The average NICD/full-length PS-1 ratio from two independent experiments is shown (bottom panel). Bar denotes S.D. Wt, wild type; FL, full-length.

 
Both PS-1 D257A and D385A Mutants Fail to Cleave Notch in Their Endoproteolyzed Forms, but the PS-1 D385A Mutant Can Restore Its {gamma}-Secretase Activity with the Compensatory Overexpression of Normal CTF—Both aspartate residues, Asp-257 and Asp-385, are required for {gamma}-secretase activity. There are controversial reports debating whether the defects of D257A and D385A mutants are the result of a change in their catalytic domains or a lack of endoproteolytic activity (12–18). To demonstrate directly that the defects of PS-1 D257A and D385A mutants are caused by changes in their catalytic sites and also to analyze the different effects of the PS-1 NTF and CTF, we constructed normal and mutant NTFs and CTFs tagged with GFP at the N terminus as follows: wt NTF (1–298 amino acids), wt CTF (299–467 amino acids), D257A NTF (1–298 amino acids in which the aspartate in residue 257 is point mutated to alanine), D385A CTF (299–467 amino acids in which the aspartate in residue 385 is point mutated to alanine). The overexpression of either wt NTF or wt CTF alone (Fig. 2A, lanes 4 and 5, respectively) was not sufficient to cleave Notch, but overexpression of wt NTF and wt CTF together (Fig. 2A, lane 6) cleaved Notch to the same extent as full-length PS-1 (Fig. 2A, lane 3), indicating that the NTF and CTF constructs functioned as expected. When we combined D257A NTF and D385A CTF with their normal counterparts, wt CTF and wt NTF, respectively (Fig. 2A, lanes 7 and 8), Notch was not cleaved, indicating that the defects of full-length PS-1 D257A and D385A are caused by changes in their catalytic domains rather than the lack of endoproteolysis. Interestingly, the defects of full-length PS-1 D385A could be restored by overexpressing its normal counterpart, wt CTF (Fig. 2A, lane 10), whereas the defect of D257A could not be restored by wt NTF (Fig. 2A, lane 9). Along the same line, it is worth noting that the defects in cleaving Notch by the C410Y mutant was also restored by overexpressing wt CTF but not wt NTF (Fig. 2A, lanes 13 and 12, respectively). Our results suggest that the defects of NTF and CTF regional mutations can be restored differentially by overexpressing their normal PS-1 fragments.

Endoproteolysis of PS-1 D257A and D385A Mutants Depends on the Presence of Endogenous PS-1/2—To measure the level of endoproteolysis of D257A and D385A mutants, we transiently transfected both mutants to PS-1/2-/- MEF cell lines that had neither endogenous PS-1 nor PS-2. In PS-1/2-/- cell lines, a generated endoproteolyzed form of PS-1 was detected with a specific antibody against their CTF. The endoproteolyzed form of PS-1 D257A and D385A was barely detectable compared with that of wt PS-1 (Fig. 2B, lanes 1, 2, and 5). The compensatory overexpression of NTF or CTF did not induce any significant increase in the endoproteolysis of PS-1 D257A and D385A (Fig. 2B, lanes 3, 4, 6, and 7). Therefore, the restoration of {gamma}-secretase activity in PS-1 D385A by the compensatory overexpression of normal CTF may result from the direct binding to the Asp-257 NTF portion of uncleaved PS-1 D385A rather than from increased endoproteolysis. The weak endoproteolysis of PS-1 D257A and D385A mutants in PS-1/2-/- MEF cells was more apparent in PS-1/2+/+ cells, suggesting that the degree of endoproteolysis of PS-1 D257A and D385A mutants might be affected by the presence of functional {gamma}-secretase (Fig. 2C).

The Minimal Domain of Wild Type CTF Maintaining the {gamma}-Secretase Activity Is Located between Residues 349 and 467, and Notch Cleavage Can Be Abolished with D385A CTF and Its 349–467 Fragment in a Dose-dependent Manner—As shown above, PS-1 is processed endoproteolytically, generating NTF and CTF. Laudon and co-workers (42) have reported that the overexpression of both wt NTF and CTF, but not necessarily full-length PS-1, is sufficient to cleave Notch, which is in agreement with our current observations. To understand the roles of NTF and CTF more precisely and map the minimal regions responsible for {gamma}-secretase activity, we constructed several deleted NTFs and CTFs tagged with GFP at their N termini, as illustrated in Fig. 3A. We confirmed the expression of each construct by Western blot analysis (Fig. 3B). Wild type CTF and deleted NTF and CTF fragments migrated as we expected, but wt NTF was incorporated mainly into HMW aggregates. Fragments B and E as well as CTF fragment A were slightly heavier than we expected, possibly because of the phosphorylation of residue 346, a known target site of protein kinase C (43). When we coexpressed one of the deleted NTFs with wt CTF in PS-1/2-/- MEFs, none of the deleted NTFs showed Notch cleavage, suggesting that an almost full conformation of wt NTF is required for {gamma}-secretase activity (Fig. 3C). Nevertheless, we found that the minimal domain of wt CTF responsible for generating {gamma}-secretase activity is located between residues 349 and 467 of PS-1 (Fig. 3D, F fragment). Neither fragment E (residues 299–417) nor fragment D (residues 418–467) was sufficient to cleave Notch.



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 2.
Both PS-1 D257A and D385A mutants fail to cleave Notch even in their endoproteolyzed forms, and the PS-1 D385A mutant can restore {gamma}-secretase activity with the compensatory overexpression of wt CTF. A, PS-1 endoproteolyzed forms tagged with GFP in their N termini, wt NTF (amino acids 1–298), wt CTF (299–467), D257A NTF (1–298, in which Asp-257 is point mutated to alanine), D385A CTF (299–467, in which Asp-385 is point mutated to alanine) were generated and transfected with several combinations into the PS-1/2-/- cells as indicated. When D257A NTF and D385A CTF were combined with its normal counterpart, Notch cleavage did not occur (lanes 7 and 8). Full-length (FL) D385A defective in Notch cleavage restored its ability by compensatory overexpression of wt CTF, but full-length D257A could not restore it by wt NTF (lanes 9 and 10). The same pattern was observed in the defective spontaneous PS-1 mutant, C410Y (lanes 11–13). The expression levels of the different PS-1 variants were analyzed with anti-PS-1 Ab (bottom panel). B, PS-1 constructs were transfected with NTF or CTF into PS-1/2-/- cells and blotted by specific antibody against the C-terminal fragment (450–467) of PS-1. Lane 8 shows lysates from the PS-1/2+/+ cells overexpressing wt PS-1. C, PS-1 constructs were transfected to PS-1/2+/+ cells and blotted with specific antibody against the C-terminal fragment (450–467) of PS-1.

 
To define the factors affecting the PS-1/{gamma}-secretase activity in CTF, especially fragments E and F, we transfected wt NTF and a different CTF fragment in PS-1/2-/- MEFs and performed immunoblot analysis as shown in Fig. 4. Noticeable decreases in Nicastrin and Notch binding to PS-1 were observed in cells transfected with wt NTF and fragment E (residues 299–417), whereas no decrease in binding in cells transfected with wt NTF and fragment F (residues 349–467) was seen, indicating that the C terminus of PS-1 (amino acids 418–467) might be important for Nicastrin and/or Notch binding (Fig. 4A). The binding of Nicastrin and Notch in wt NTF and D385A CTF-transfected cells was observed (second lane), indicating once more that the defects of PS-1 D385A in Notch cleavage may be caused by changes in the catalytic sites but not by the lack of binding to Nicastrin and Notch or endoproteolysis, as shown above.

As the coexpression of wild type CTF restored the defective Notch cleavage of full-length PS-1 D385A, we predicted that the overexpression of D385A CTF could abolish {gamma}-secretase activity driven by wt PS-1. We coexpressed D385A CTF with wt and various mutant PS-1s in PS-1/2-/- MEFs and analyzed Notch cleavage patterns. Protein sample loadings on each lane were confirmed using the direct blue 71 staining kit and also quantified by reprobing membranes with anti-tubulin Ab, as shown in Fig. 5A. As shown in Fig. 5, the coexpression of D385A CTF (fragment A, residues 299–467) clearly decreased the Notch cleaving ability of both wt and mutant PS-1s. Furthermore, the expression of both D385A CTF (fragment A) and minimal D385A CTF (fragment F, residues 349–467) also showed inhibitory effects on Notch cleaving ability of wt PS-1 in a dose-dependent manner (Fig. 5B). Fragments C and E showed minor inhibitory effects on Notch cleavage (~10% reduction). We also coexpressed D385A CTF or D257A NTF with various mutant PS-1s in PS-1/2-/- MEFs and analyzed the cleavage patterns of Notch. As shown in Fig. 5C, the coexpression of D385A CTF clearly decreased the Notch cleaving ability of both wt and mutant PS-1s. Interestingly, Notch cleavage was also decreased by the coexpression of D257A NTF, in contrast with the restoration experiments shown above.



View larger version (39K):
[in this window]
[in a new window]
  		FIG. 3.
The minimal domain of wt CTF maintaining {gamma}-secretase activity is mapped between residues 349 and 468. A and B, schematic diagram of PS-1 deletion fragments and expression of each fragment in PS-1/2+/+ cells. Different intensities between each fragment may be a consequence of their stability in cells or of different integration into HMW aggregates. C, NTF deletion fragments were transfected to PS-1/2-/- cells with its counterpart, wt CTF and its substrate, Notch. To investigate the minimal NTF domain, Notch cleavage activity was measured by Western blot analysis with anti-NICD antibody. None of these deletion fragments showed Notch cleavage, suggesting that an almost full conformation of wt NTF might be required for {gamma}-secretase activity. D, to investigate the minimal CTF domain, CTF deletion fragments were transfected to PS-1/2-/- cells with wt NTF and Notch, and Western blot analysis was done using anti-NICD Ab. Note that fragment F (amino acids 350–467) was sufficient for cleaving Notch, but neither fragments 299–417 (E) nor 418–467 (D) alone was sufficient to cleave Notch.

 
The Formation of HMW Aggregates Is Required but Not Sufficient for Notch Cleavage—While performing SDS-PAGE for Western blot analysis, we noticed dramatic differences in migration rates of various NTFs and CTFs (Fig. 3B). In fact, it has been reported that PSs have a strong tendency to form aggregates (100–250 kDa), even in SDS-PAGE (44). In the present experiments, PS-1 HMW aggregates were detected with GFP Ab bound to full-length PS-1 at the C terminus of the GFP epitope (Fig. 6A, second lane) and also bound to PS-1 NTF at the GFP C terminus (Fig. 6A, third lane). PS-1 and NTF tagged with GFP equally form HMW aggregates, but NTF aggregates are denser than that of full-length PS-1. GFP-CTF, however, did not form such aggregates (Fig. 6A, fourth lane). Even when both NTF and CTF were coexpressed, overexpressed CTF did not show noticeable shifts to HMW aggregates (Fig. 6A, fifth lane).

To know whether the formation of HMW aggregates relies on the presence of PS-1 and PS-2, we compared four MEFs displaying different levels of endogenous PS-1 and PS-2. The most dramatic formation of HMW aggregates was observed in PS-1/2+/+ MEFs, although we also observed weak HMW aggregates formation in both PS-1 -/- and -2 -/- MEFs (data not shown). However, we hardly detected the presence of HMW aggregates in PS-1/2-/- MEFs, suggesting that the factor(s) responsible for the formation of HMW aggregates could be missing or impaired in PS-1/2-/- MEFs.



View larger version (53K):
[in this window]
[in a new window]
  		FIG. 4.
Decreased binding of Nicastrin and Notch to CTF E fragment result in Notch cleavage defects. Nicastrin-V5, Pen2-HA, Notch-myc, and GFP-tagged PS-1 NTF and CTF fragments were transfected into PS-1/2-/- cells, and immunoprecipitation analysis was done as described under "Materials and Methods." A, whole cell lysates were immunoprecipitated with GFP Ab and immunoblotted with Ab as described. The PS-1 349–468 fragment (E) showed a noticeable decline in binding to the {gamma}-secretase component, Nicastrin and its substrate, Notch. B, Western blot analysis of whole cell lysates using anti-GFP, -V5, and -myc Abs showed expression of each construct, PS-1 fragments, Nicastrin, and Notch, respectively. Cells transfected with PS-1 C410Y show a denser Nicastrin band, which may be the result of its intact configuration. Asterisks indicate nonspecific bands.

 
As NTF and CTF showed differences in integration with HMW aggregates, we tagged the full-length PS-1 either at the N terminus (GFP-PS-1) or the C terminus (C-GFP-PS-1) and transfected them into PS-1/2+/+ cells. We then performed Western blot analysis using a peptide antibody specific for GFP. We used 8% SDS-PAGE and ran it 30 min longer than usual to get the best resolution possible. HMW bands were not observed in control MEFs (Fig. 6B, first lane). However, three major HMW bands were detected in the cells transfected with PS-1 (both in GFP-PS-1 and C-GFP-PS-1), with noticeable differences in their intensities. As shown in Fig. 6B, second and third lanes, C-GFP-PS-1 showed a stronger bottom band, whereas GFP-PS-1 did not (Fig. 6B). Their molecular masses were estimated to be approximately >200 kDa. Increasing {beta}-mercaptoethanol and SDS content, as well as boiling duration, did not affect the results (data not shown). However, the treatment at 37 °C dissociated the HMW aggregates, and some GFP-NTF was visible, whereas 65 °C did not break HMW aggregates (Fig. 6D). Similarly, the treatment at 95 °C dissociated the HMW aggregates and brought down upper HMW bands to lower HMW bands, but GFP-NTF was not visible. These results strongly suggest that HMW aggregates are "temperature-sensitive" as also reported by De Strooper et al. (44). Only when lysates were treated with the powerful detergent guanidine HCl (6 M), HMW bands became much weaker (data not shown), suggesting that components of HMW aggregates bind very tightly.

Because the formation of HMW aggregates depended on the overexpression of NTF, we investigated the region responsible for HMW aggregate formation by using all deleted NTF fragments. Only full-length NTF formed strong HMW aggregates (see Fig. 3B). In contrast to fragments N131 and N200, fragment N260 formed aggregates, indicating that the region comprised by amino acids 200–260 plays a major role in aggregate formation. Fragment 131–298, which comprises the 200–260 amino acid region, also formed aggregates but was weaker than fragment N260, indicating that amino acids 1–130 from the N terminus play a minor role in strong HMW aggregate formation.

Because the overexpression of exogenous PS-1 is required for generating {gamma}-secretase activity in PS-1/2-/- MEFs, we overexpressed various mutant PS-1s tagged with GFP at the N terminus. When we performed Western blot analysis for various PS-1 mutants, defective PS-1 mutants, including PS-1 D257A, D385A, and C410Y, showed HMW aggregate formation, although the HMW bands were slightly shifted and had different intensities (Fig. 6D). Therefore, the formation of HMW aggregates appears to be required for Notch cleavage, although aggregate formation alone is not sufficient for cleavage to occur.

Exogenous CTF was Coimmunoprecipitated with Uncleaved PS-1 D385A, whereas NTF Was Not with PS-1 D257A—Based on our observations, as well as published reports by others, we proposed a new model for PS-1 conformation (Fig. 7). According to our model, the critical aspartate residues, Asp-257 and Asp-385, in uncleaved PS-1 do not face each other sterically. Only the Asp-257 portion of the PS-1 NTF is exposed outward, whereas the Asp-385 portion of the PS-1 CTF is sterically hindered inward. To prove our model, we performed immunoprecipitation assays. We cotransfected full-length PS-1 D257A and D385A mutants tagged with FLAG, and NTF and CTF tagged with GFP, respectively, to PS-1/2-/- MEFs as shown in Fig. 8. We immunoprecipitated protein extracts with anti-FLAG Ab and blotted with anti-GFP Ab. Exogenous CTF was coimmunoprecipitated with uncleaved PS-1 D385A, whereas NTF was not with PS-1 D257A (Fig. 8), directly addressing our hypothesis that only the Asp-257 portion of the PS-1 NTF is exposed outward, whereas the Asp-385 portion of the PS-1 CTF is sterically hindered inward.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this report we have demonstrated the following: 1) Notch cleavage depends on the presence of endogenous PS-1 and PS-2 and on the type of PS-1 mutants; 2) both PS-1 D257A and D385A mutants fail to cleave Notch, even in their endoproteolyzed forms, but only the PS-1 D385A mutant, under compensatory overexpression of normal CTF, can restore {gamma}-secretase activity; 3) the minimal domain required for wt CTF to maintain {gamma}-secretase activity was determined to be between residues 349 and 467, and Notch cleavage can be abolished with D385A CTF and its 349–467 fragment in a dose-dependent manner.

Because PS-1 is a major component of the {gamma}-secretase complex, it is not surprising to find that PS-1 mutants affect the {gamma}-secretase catalytic activity, endoproteolytic process, association with other {gamma}-secretase components, and the binding affinity to its substrates. However, when we performed restoration experiments for the defective PS-1 mutants, it was of great interest to find that their endoproteolyzed forms, NTF and CTF, had different effects on restoring the defective PS-1 mutants with regard to Notch cleavage. To date, NTF is known to interact with Pen-2, whereas CTF interacts with Nicastrin (45) and contains a caspase recognition site (43). The CTF of PS-2 has been shown to affect apoptosis (46), but the different biological functions and regulation of NTF and CTF have not been clearly documented yet. We noticed dramatic differences between PS-1 NTF and CTF, both with respect to restoring Notch cleavage ability of defective PS-1 mutants as well as with different running patterns in SDS-PAGE. When the PS-1 mutation that prevents Notch cleavage is located at the NTF (i.e. D257A), the overexpression of normal wt NTF could not restore {gamma}-secretase activity. In contrast, when the PS-1 mutation that prevents Notch cleavage is located at the CTF (i.e. D385A and C410Y), the overexpression of normal CTF was shown to restore {gamma}-secretase activity.



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 5.
The cleavage of Notch can be abolished with D385A CTF and its 349–468 fragment in a dose-dependent manner. A, PS-1 D385A CTF (amino acids 299–467, in which Asp-257 is point mutated to alanine) and its deletion fragments were transfected into PS-1/2-/- cells with Notch and wt PS-1. Cell lysates were blotted with anti-NICD Ab (top panel) and reprobed with anti-tubulin Ab (middle panel). For the NICD/tubulin ratio, three independent transfections were done, and a quantification of the protein bands in three different blots was performed using the TINA 2.09 software program. D385A CTF (D385A, A) and D385A fragment 349–468 (D385A, F) inhibited NICD generation (~30% reduction), and CTF fragments C and E moderately inhibited (~10% reduction). Bar denotes S.D. B, the expression of both D385A CTF (fragment A) and minimal Asp-385 CTF (fragment F, residues 349–467) also showed inhibitory effects on Notch cleaving ability of wt PS-1 in a dose-dependent manner. C, D385A CTF not only inhibited wt PS-1 but also several PS-1 mutants. D257A NTF also inhibited the formation of NICD.

 
We then investigated the mechanism by which overexpression of either wt NTF or CTF could differentially restore its defective PS-1 mutant counterpart. Based on our observations as well as published reports by others, we proposed a new model for PS-1 conformation. According to our model, the critical aspartate residues, Asp-257 and Asp-385, in uncleaved PS-1 do not face each other sterically (Fig. 7A). When other components of a {gamma}-secretase complex (i.e. Aph-1, Nicastrin, Pen-2) bind to PS-1, it induces conformational changes in PS-1. This results in the formation of a functional bi-aspartate bond between residues Asp-257 and Asp-385, which is critical for active {gamma}-secretase activity (Fig. 7B). Based on these results, we proposed a model in which only the Asp-257 portion of the PS-1 NTF is exposed outward, whereas the Asp-385 portion of the PS-1 CTF is sterically hindered inward (Fig. 7). For optimal {gamma}-secretase activity to occur, heterodimers of NTF and CTF should be formed at the core of the enzyme so that regions Asp-257 and Asp-385 can face each other. When we co-overexpress wt CTF with the full-length PS-1 D385A mutant, residue Asp-385 in wt CTF can form a bi-aspartate bond with the exposed residue Asp-257 in the uncleaved full-length PS-1 D385A mutant (Fig. 7C). The newly formed bi-aspartate bond enables full {gamma}-secretase activity. Along this line, it is of interest to note that the PS-1 C410Y mutant, having intact Asp-257 and Asp-385 residues, but defective in its ability to undergo endoproteolysis, could restore {gamma}-secretase activity when wt CTF was co-overexpressed. Our results showed that PS-1 C410Y could bind to Nicastrin and form HMW aggregates. This suggests that a mutation in C410Y affects the conformation of PS-1, which results in defective endoproteolysis. However, as in the case of the PS-1 D385A mutant, the co-overexpressed wt CTF can get access to the Asp-257 portion of PS-1 C410Y and form the functional bi-aspartate bond for {gamma}-secretase activity. In contrast, when we coexpressed wt NTF with full-length D257A, wt NTF could not obtain access to the Asp-385 portion because it is sterically hindered inward (Fig. 7D). Even though both PS-1 D257A and D385A mutants showed barely detectable levels of endoproteolysis in PS-1/2-/- MEFs, we could observe large differences during restoration experiments, possibly because of their different conformations and accessibility to overexpressed normal counterparts. This model is supported further by the results showing that compensatory overexpression of NTF or CTF did not induce any significant increase in the endoproteolysis of PS-1 D257A and D385A in PS-1/2-/- MEFs. Also, exogenous PS-1 CTF coimmunoprecipitated with uncleaved PS-1 D385A in PS-1/2-/- MEFs, whereas NTF did not with PS-1 D257A. Therefore, the restoration of {gamma}-secretase activity in PS-1 D385A by the compensatory overexpression of normal CTF might be the result of its direct binding to the Asp-257 NTF portion of uncleaved PS-1 D385A rather than increased endoproteolysis.



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 6.
Formation of PS-1 HMW aggregates appears to be required for Notch cleavage, although aggregate formation alone is not sufficient for cleavage to occur. A, full-length PS-1, NTF, and CTF tagged with GFP at their N termini were transfected into PS-1/2+/+ cells, and Western blot analysis was carried out with anti-GFP antibody. The asterisk indicates a putative dimer of NTF. The NTF monomer band was overlapped by a nonspecific band. B, wt PS-1 tagged with GFP at the N termini (N-GFP-PS-1) or C termini (C-GFP-PS-1) was transfected into PS-1/2+/+ cells. Western blot analysis was done with anti-GFP antibody. GFP-tagged PS-1 showed strong aggregation at molecular masses higher than 200 kDa. C, GFP-PS-1-transfected cell lysates were denatured for 2 min at 95 °C (left lane), 30 min at 65 °C (center lane), or 30 min at 37 °C (right lane). A decrease in HMW aggregates and compensatory increase of the NTF band in the right lane can be observed. D, wt and various PS-1 constructs were transfected to PS-1/2+/+ cells, and Western blot analysis was done with anti-GFP. The same PS-1 constructs and wt were transfected to PS-1/2-/- cells with Notch.

 
In addition to providing conformational information in PS-1, we narrowed down the domains in PS-1, which are critical for {gamma}-secretase activity. In these experiments we observed that {gamma}-secretase activity of wt PS-1 required full conformation of NTF (residues 1–298), but we managed to narrow down the minimal functional domain of CTF to the region between residues 349 and 468. Although Hebert et al. (47) showed that TM1–2 (amino acids 101–153) in PS-1 formed a heterodimer with wt CTF by using a yeast two hybrid assay, our results indicated that 1–260 NTF plus wt CTF was not sufficient to cleave Notch. When we performed immunoprecipitation experiments, a noticeable decrease in Nicastrin and Notch binding to PS-1 was observed in PS-1/2-/- cells overexpressing wt NTF (residues 1–298) and the CTF fragment E (residues 299–417), which lacks the 418–468 C terminus. Recently, Bergman et al. (48) reported that the extreme C terminus of PS-1 is essential for endoproteolysis and {gamma}-secretase activity. Therefore, taken together with past studies, our results strongly suggest that the C terminus of PS-1 (residues 418–468 or shorter) might be important for Nicastrin binding and subsequent {gamma}-secretase activity.

Whether or not the PS-1 itself is presenilinase is a longstanding puzzle in this field. To date, it is known that Pen-2 is a regulator of PS-1 endoproteolysis (23, 25). Bergman et al. (48) reported that the extreme C terminus of PS-1 is also essential for endoproteolysis. In this report, we added clues that PS itself indeed can be a presenilinase or a major component of presenilinase. When we blotted protein samples with the antibody specific for the PS-1 C terminus, full-length PS-1 D257A, D385A, and C410Y mutants showed barely detectable levels of endoproteolysis in PS-1/2-/- MEFs. However, it is of great interest to note that the endoproteolysis of PS-1 D257A, D385A, and C410Y mutants was clearly observed in PS-1/2+/+ cells, possibly because of the presence of endogenous functional PS complexes. The overexpressed Pen-2, or any combinations of major components in the {gamma}-secretase complex, could not restore the defective {gamma}-secretase activity of PS-1 D257A under the compensatory overexpression of normal NTF.2 This indicates that the conformational change of PS-1 D257A, induced by overexpressed Pen-2 itself, could not be sufficient to allow wt NTF access to the CTF of uncleaved PS-1 D257A. Alternatively, in addition to Pen-2, there may be another factor(s) required for the conformational change of PS-1. In contrast, while restoring PS-1 D385A, overexpressed wt CTF can form the functional {gamma}-secretase complex with the Asp-257 portion of PS-1 D385A but does not significantly increase endoproteolysis. These data suggest that the catalytic sites of presenilinase and {gamma}-secretase can be spatially separated. Taken together, the functional assembly between NTF and CTF may be important for both presenilinase and {gamma}-secretase activity.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 7.
Proposed model illustrating the mechanism for {gamma}-secretase function. A, diagram of PS-1 (left) with a simplified side view (upper right) and top view (bottom right). Aspartates at residues 257 and 385 are sterically hindered to each other in uncleaved PS-1. B, diagram showing active {gamma}-secretase. By binding to other {gamma}-secretase components, PS-1 is endoproteolytically processed, generating NTF and CTF. As a result, aspartates at residues 257 and 385 are faced against each other to form a bi-aspartate bond. C, diagram illustrating the restoring mechanism of PS-1 D385A defective in its ability to cleave Notch by coexpressing wt CTF. Aspartate at residue 385 of wt CTF could form the bi-aspartate bond with aspartate at residue 257 of uncleaved PS-1 D257A. D, diagram illustrating the lack of restoration of Notch cleavage-defective PS-1 D257A by coexpressing wt NTF. Aspartate at residue 257 of wt NTF cannot access aspartate at residue 385 of uncleaved PS-1 D385A. NCT, Nicastrin.

 
Nonetheless, the endoproteolysis of PS-1 may not be an absolute prerequisite for {gamma}-secretase activity. The presence of {gamma}-secretase activity has been shown for the noncleaved FAD-associated PS-1 {Delta}E9 mutants and artificial M292D mutants (49). In this report, we also clearly demonstrated that the artificially cleaved forms of D257A NTF and D385A CTF showed an inability to cleave Notch, even when coexpressed with its cleaved counterparts, wt CTF and NTF, respectively, which rules out the possibility that Notch cleavage defects caused by PS-1 D257A and D385A are caused by a lack of endoproteolysis. Likewise, the artificial Y288N mutant has been shown to be endoproteolytically processed, but it could not reconstitute {gamma}-secretase in PS-null cells (50). PS-1 mutated at Asp-257 or Asp-385 has been shown to restore Pen-2 expression and Nicastrin glycosylation (15). Our results showed that M146V/D257A double mutants had defects in cleaving Notch, strongly indicating that the presence of both Asp-257 and Asp-385 residues is required for full {gamma}-secretase activity. Therefore, both the assembly of Nicastrin, Aph-1, and Pen-2 into PS, and the formation of the PS-1 bi-aspartate bond at the core of {gamma}-secretase complexes are critical for {gamma}-secretase activity. The conformational changes of PS-1 mutants resulting from a point mutation may be important to determine {gamma}-secretase activity. Different PS-1 mutants may have unequal access to Nicastrin and Pen-2, and the formation of the PS-1 bi-aspartate can also be differentially affected depending on the type or position of the PS mutation.



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 8.
Exogenous CTF is coimmunoprecipitated with uncleaved PS-1 D385A, whereas NTF is not with PS-1 D257A. PS-1/2-/- MEFs were transfected with full-length PS-1 D257A or D385A mutants tagged with FLAG, and with NTF or CTF tagged with GFP as indicated on the top of figure. Protein extracts were immunoprecipitated (IP) with anti-FLAG Ab and blotted with anti-GFP Ab (top panel). To show the expression levels of CTF and NTF, whole cell lysates before immunoprecipitation were blotted with anti-GFP Ab (bottom panel).

 
As shown by past work as well as our present results, PS-1 mutants defective in its own and Notch endoproteolysis, including PS-1 D257A, D385A, {Delta}Exon 9, C410Y, M292D, and Y288D, are still able to become incorporated into HMW complexes (15). Therefore, the formation of HMW aggregates appears to be required for Notch cleavage, although aggregate formation alone is not sufficient for cleavage to occur.

Aside from restoration experiments, noticeable differences between NTF and CTF could also be seen in the formation of HMW complexes while running SDS-PAGE. Most importantly, wt NTF formed a band not only at its estimated molecular mass, but also at ~200–300 kDa above. Wild type CTF, on the other hand, ran strictly in accordance to its estimated molecular mass. As described under "Results" (Fig. 3B), the NTF region between residues 200 and 260 is critical for the formation of HMW aggregates. Many others have described the formation of PS-1 HMW complexes under different conditions as well as the copresence of {gamma}-secretase components; however, in those studies, it was hard to detect the PS-1 HMW complex (45, 51). The major components of the {gamma}-secretase complex include PS-1, Nicastrin, Aph-1, and Pen-2 (10, 25), whose summed molecular masses amount to ~150–200 kDa. This is smaller than the size of the active {gamma}-secretase complex predicted by gel filtration or glycerol velocity separation, which might exceed ~250–1,000 kDa (53). The coexistence of the major components in HMW complexes has been demonstrated with the use of RNA interference experiments and glycerol velocity separation (10, 20, 24, 54).

Because of the discrepancy between estimated and observed molecular masses, different stoichiometries for four major components and the presence of unidentified limiting factors were suggested. Along this line, Yu et al. (7) identified {beta}-catenin as a component of a ~250-kDa complex by velocity gradient centrifugation and coimmunoprecipitation, and Gu et al. (52) showed the presence of at least three complexes, including the high mass (~670 kDa) heteromeric complexes that are associated with the highest {gamma}-secretase specific activity. Our results demonstrated that the formation of PS-1 HMW aggregates clearly depends on the presence of PS-1 and PS-2. The most intense formation of PS-1 HMW aggregates could be observed in PS-1/2+/+ MEFs, and the least intense in PS-1/2-/- MEFs. Consistent with the results of Gu et al. (52), we also observed at least three major HMW aggregates, although they showed different intensities depending on the location of an exogenous tag.

Our new contributions to this field include the description of the different effects of NTF and CTF on the formation of HMW aggregates. Only the overexpressed NTF forms HMW aggregates, which also depend on the presence of endogenous PS-1/2, as in the case for full-length PS-1. The presence of strong HMW aggregates in PS-1/2+/+ cells strongly overexpressing wt NTF and N260 indicates that NTF either forms homodimers or multimers or, alternatively, binds very tightly to the component(s) in cells and that the major domain responsible for the formation of HMW aggregates is located between residues 200 and 260. In contrast to the strong HMW aggregates of wt NTF, wt CTF only formed a band at its estimated molecular mass. This indicates that wt CTF is likely to integrate into the {gamma}-secretase complex less tightly than wt NTF and/or remains only as a heteromer with NTF. It is interesting to note that NTF contains more transmembrane domains than CTF, which may provide different binding affinities to the components of the {gamma}-secretase complex. In a physiological context, the NTF HMW aggregates could contribute more than CTF to form the active {gamma}-secretase complex, either by binding tighter to {gamma}-secretase complex components or by forming NTF-NTF and NTF-CTF bonds with different stoichiometry. Interestingly, Schroeter et al. (18) reported that photoaffinity probes could cross-link both NTF/CTF and NTF/NTF dimers but not CTF/CTF dimers at the core of the {gamma}-secretase enzyme. In this regard, the inability of PS-1 D257A to restore Notch cleavage can result from the strong homodimeric formation of NTFs. Hence, we could not rule out the possibility that NTF has higher affinity to NTF than to CTF. If this were the case, the overexpressed wt NTFs would mostly form homodimers with D257A NTFs, resulting in a defective {gamma}-secretase complex in our rescuing experiments. In contrast, the overexpressed CTFs may form a functional heterodimer with the NTF of PS-1 D385A.

When we overexpressed the minimal D385A CTF (residues 349–468), {gamma}-secretase activity, as determined by Notch cleavage, of both wild type and pathogenic PS-1, was effectively blocked in a dose-dependent manner. Interestingly, D257A NTF also effectively blocked the Notch-cleaving {gamma}-secretase activity of both wild type and pathogenic PS-1. As discussed above, wt NTF and CTF function differently on restoring Notch cleavage of defective PS-1, which leads to the question of how both D257A NTF and D385A are able to block efficiently the {gamma}-secretase activity on Notch cleavage. All functional wt PS-1 and pathogenic mutants undergo endoproteolysis. Both D257A NTF and D385A CTF compete with functional NTFs and CTFs in cells once they are endoproteolytically processed and thereby, can reduce {gamma}-secretase activity of wild type and mutant PS-1. Although more PS-1 mutants should be analyzed, and factors differently affecting NTF and CTF need to be identified, our data strongly indicate that the CTF integrates less strongly into the PS-1 containing HMW complexes than does NTF and that the D385A CTF can be used as a valuable tool for inhibiting {gamma}-secretase activity. Because many spontaneously occurring PS-1 mutants showed different specificities and affinities to its substrates, the profiling of PS-1 mutants for the cleavage of its substrates, including amyloid precursor protein and Notch, should be carefully carried out to elucidate their pathogenic effects and to understand their biological roles in the progression of AD.


   FOOTNOTES
 
* This study was supported by Grant 0405-NS01-0704-0001 from the Biomedical Brain Research Center, Ministry of Health and Welfare, Republic of Korea (to K. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

To whom correspondence should be addressed: College of Pharmacy, Chonnam National University, Bldg. 1-211, 300 Yongbong-dong, Gwangju 500-757, Korea. Tel.: 82-62-530-2937; Fax: 82-62-530-2949; E-mail: koskim{at}chonnam.ac.kr.

1 The abbreviations used are: AD, Alzheimer disease; Ab, antibody; CTF, C-terminal fragment; FAD, familial Alzheimer disease; GFP, green fluorescent protein; HA, hemagglutinin; HMW, high molecular weight; MEF, mouse embryonic fibroblast; NICD, Notch intracellular domain; NTF, N-terminal fragment; PS, presenilin; wt, wild type. Back

2 H. Kim and K. Kim, unpublished results. Back


   ACKNOWLEDGMENTS
 
We are greatly indebted to Dr. De Strooper for the gift of four PS MEF cell lines, to Dr. Takeshi Iwatsubo for the HA-hpen-2, and to Dr. Jung-Kap Choi for providing Direct Blue 71 staining kits (EZ BiopaQ, Korea).



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Rubinsztein, D. C. (1997) Prog. Neurobiol. 52, 447-454[CrossRef][Medline] [Order article via Infotrieve]
  2. Czech, C., Tremp, G., and Pradier, L. (2000) Prog. Neurobiol. 60, 363-384[CrossRef][Medline] [Order article via Infotrieve]
  3. Cruts, M., and Van Broeckhoven, C. (1998) Hum. Mutat. 11, 183-190[CrossRef][Medline] [Order article via Infotrieve]
  4. Rogaeva, E. A., Fafel, K. C., Song, Y. Q., Medeiros, H., Sato, C., Liang, Y., Richard, E., Rogaev, E. I., Frommelt, P., Sadovnick, A. D., Meschino, W., Rockwood, K., Boss, M. A., Mayeux, R., and St. George-Hyslop, P. (2001) Neurology 28, 621-625
  5. Chen, F., Gu, Y., Hasegawa, H., Ruan, X., Arawaka, S., Fraser, P., Westaway, D., Mount, H., and St. George-Hyslop, P. (2002) J. Biol. Chem. 277, 36521-36526[Abstract/Free Full Text]
  6. Kovacs, D. M., Fausett, H. J., Page, K. J., Kim, T. W., Moir, R. D., Merriam, D. E., Hollister, R. D., Hallmark, O. G., Mancini, R., Felsenstein, K. M., Hyman, B. T., Tanzi, R. E., and Wasco, W. (1996) Nat. Med. 2, 224-229[CrossRef][Medline] [Order article via Infotrieve]
  7. Yu, G., Chen, F., Levesque, G., Nishimura, M., Zhang, D., Levesque, L., Rogaeva, E., Xu, D., Liang, Y., Liang, Y., Duthie, M., St. George-Hyslop, P. H., and Fraser, P. E. (1998) J. Biol. Chem. 273, 16470-16475[Abstract/Free Full Text]
  8. Podlisny, M. B., Citron, M., Amarante, P., Sherrington, R., Xia, W., Zhang, J., Diehl, T., Levesque, G., Fraser, P., Haass, C., Koo, E. H., Seubert, P., St. George-Hyslop, P., Teplow, D. B., and Selkoe, D. J. (1997) Neurobiol. Dis. 17, 325-337
  9. Shirotani, K., Edbauer, D., Capell, A., Schmitz, J., Steiner, H., and Haass, C. (2003) J. Biol. Chem. 278, 16474-16477[Abstract/Free Full Text]
  10. Kimberly, W. T., LaVoie, M. J., Ostaszewski, B. L., Ye, W., Wolfe, M. S., and Selkoe, D. J. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 6382-6387[Abstract/Free Full Text]
  11. Knappenberger, K. S., Tian, G., Ye, X., Sobotka-Briner, C., Ghanekar, S. V., Greenberg, B. D., and Scott, C. W. (2004) Biochemistry 43, 6208-6218[CrossRef][Medline] [Order article via Infotrieve]
  12. Yu, G., Chen, F., Nishimura, M., Steiner, H., Tandon, A., Kawarai, T., Arawaka, S., Supala, A., Song, Y. Q., Rogaeva, E., Holmes, E., Zhang, D. M., Milman, P., Fraser, P., Haass, C., and St. George-Hyslop, P. (2000) J. Biol. Chem. 275, 27348-27353[Abstract/Free Full Text]
  13. Levitan, D., Lee, J., Song, L., Manning, R., Wong, G., Parker, E., and Zhang, L. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 12186-12190[Abstract/Free Full Text]
  14. Xia, X., Wang, P., Sun, X., Soriano, S., Shum, W. K., Yamaguchi, H., Trumbauer, M. E., Takashima, A., Koo, E. H., and Zheng, H. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 8760-8765[Abstract/Free Full Text]
  15. Nyabi, O., Bentahir, M., Horre, K., Herreman, A., Gottardi-Littell, N., Van Broeckhoven, C., Merchiers, P., Spittaels, K., Annaert, W., and De Strooper, B. (2003) J. Biol. Chem. 278, 43430-43436[Abstract/Free Full Text]
  16. Capell, A., Steiner, H., Romig, H., Keck, S., Baader, M., Grim, M. G., Baumeister, R., and Haass, C. (2000) Nat. Cell Biol. 2, 205-211[CrossRef][Medline] [Order article via Infotrieve]
  17. Edbauer, D., Winkler, E., Regula, J. T., Pesold, B., Steiner, H., and Haass, C. (2003) Nat. Cell Biol. 5, 486-488[CrossRef][Medline] [Order article via Infotrieve]
  18. Schroeter, E. H., Ilagan, M. X. G., Brunkan, A. L., Hecimovic, S., Li, Y. M., Xu, M., Lewis, H. D., Saxena, M. T., De Strooper, B., Coonrod, A., Tomita, T., Iwatsubo, T., Moore, C. L., Goate, A., Wolfe, M. S., Shearman, M., and Kopan, R. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 13075-13080[Abstract/Free Full Text]
  19. Leem, J. Y., Vijayan, S., Han, P., Cai, D., Machura, M., Lopes, K. O., Veselits, M. L., Xu, H., and Thinakaran, G. (2002) J. Biol. Chem. 277, 19236-19240[Abstract/Free Full Text]
  20. Bergman, A., Hansson, E. M., Pursglove, S. E., Farmery, M. R., Lannfelt, L., Lendahl, U., Lundkvist, J., and Naslund, J. (2004) J. Biol. Chem. 279, 16744-16753[Abstract/Free Full Text]
  21. Crystal, A. S., Morais, V. A., Fortna, R. R., Carlin, D., Pierson, T. C., Wilson, C. A., Lee, V. M., and Doms, R. W. (2004) Biochemistry 43, 3555-3563[CrossRef][Medline] [Order article via Infotrieve]
  22. Luo, W., Wang, H., Li, H., Kim, B. S., Shah, S., Lee, H., Thinakaran, G., Kim, T., Yu, G., and Xu, H. (2003) J. Biol. Chem. 278, 7850-7854[Abstract/Free Full Text]
  23. Prokop, S., Shirotani, K., Edbaur, D., Haass, C., and Steiner, H. (2004) J. Biol. Chem. 279, 23255-23261[Abstract/Free Full Text]
  24. Lee, S., Shah, S., Li, H., Yu, C., Han, W., and Yu, G. (2002) J. Biol. Chem. 277, 45013-45019[Abstract/Free Full Text]
  25. Takasugi, N., Tomita, T., Hayashi, I., Tsuruoka, M., Nimura, M., Takahashi, Y., Thinakaran, G., and Iwatsubo, T. (2003) Nature 422, 438-441[CrossRef][Medline] [Order article via Infotrieve]
  26. De Strooper, B., Saftig, P., Craessaerts, K., Vanderstichele, H., Guhde, G., Annaert, W., Von Figura, K., and Van Leuven, F. (1998) Nature 391, 387-390[CrossRef][Medline] [Order article via Infotrieve]
  27. Naruse, S., Thinakaran, G., Luo, J. J., Kusiak, J. W., Tomita, T., Iwatsubo, T., Qian, X., Ginty, D. D., Price, D. L., Borchelt, D. R., Wong, P. C., and Sisodia, S. S. (1998) Neuron 21, 1213-1221[CrossRef][Medline] [Order article via Infotrieve]
  28. Lammich, S., Okochi, M., Takeda, M., Kaether, C., Capell, A., Zimmer, A. K., Edbauer, D., Walter, J., Steiner, H., and Haass, C. (2002) J. Biol. Chem. 277, 44754-44759[Abstract/Free Full Text]
  29. Murakami, D., Okamoto, I., Nagano, O., Kawano, Y., Tomita, T., Iwatsubo, T., De Strooper, B., Yumoto, E., and Saya, H. (2003) Oncogene 22, 1511-1516[CrossRef][Medline] [Order article via Infotrieve]
  30. Marambaud, P., Shioi, J., Serban, G., Georgakopoulos, A., Sarner, S., Nagy, V., Baki, L., Wen, P., Efthimiopoulos, S., Shao, Z., Wisniewski, T., and Robakis, N. K. (2002) EMBO J. 21, 1948-1956[CrossRef][Medline] [Order article via Infotrieve]
  31. Ni, C. Y., Murphy, M. P., Golde, T. E., and Carpenter, G. (2001) Science 294, 2179-2181[Abstract/Free Full Text]
  32. Struhl, G., and Greenwald, I. (1999) Nature 398, 522-525[CrossRef][Medline] [Order article via Infotrieve]
  33. Ye, Y., Lukinova, N., and Fortini, M. E. (1999) Nature 398, 525-529[CrossRef][Medline] [Order article via Infotrieve]
  34. 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., and Kopan, R. (1999) Nature 398, 518-522[CrossRef][Medline] [Order article via Infotrieve]
  35. Fortini, M. E. (2002) Nat. Rev. Mol. Cell. Biol. 3, 673-684[CrossRef][Medline] [Order article via Infotrieve]
  36. De Strooper, B., and Annaert, W. (2001) J. Cell Biol. 152, F17-F20[Abstract/Free Full Text]
  37. Ross, D., and Kadesch, T. (2001) Mol. Cell. Biol. 21, 7537-7544[Abstract/Free Full Text]
  38. Six, E., Ndiaye, D., Laabi, Y., Brou, C., Gupta-Rossi, N., Israel, A., and Logeat, F. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 7638-7643[Abstract/Free Full Text]
  39. LaVoie, M., and Selkoe, D. J. (2003) J. Biol. Chem. 278, 34427-34437