Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2002 Jun 17;99(13):8760–8765. doi: 10.1073/pnas.132045399

The aspartate-257 of presenilin 1 is indispensable for mouse development and production of β-amyloid peptides through β-catenin-independent mechanisms

Xuefeng Xia *, Pei Wang *,†, Xiaoyan Sun ‡, Salvador Soriano §, Wan-Kyng Shum *, Haruyasu Yamaguchi , Myrna E Trumbauer ‖, Akihiko Takashima ‡, Edward H Koo §, Hui Zheng *,†,**,‡‡
PMCID: PMC124372  PMID: 12070348

Abstract

To differentiate multiple activities of presenilin 1 (PS1), we generated transgenic mice expressing two human PS1 alleles: one with the aspartate to alanine mutation at residue 257 (hPS1D257A) that impairs the proteolytic activity of PS1, and the other deleting amino acids 340–371 of the hydrophilic loop sequence (hPS1Δcat) essential for β-catenin interaction. We show here that although hPS1Δcat is fully competent in rescuing the PS1-null lethal phenotype, hPS1D257A does not exhibit developmental activity. hPS1D257A also leads to the concurrent loss of the proteolytic processing of Notch and β-amyloid precursor protein (APP) and the generation of β-amyloid peptides (Aβ). Further, by measuring the levels of endogenous AβX-40 and AβX-42 in primary neuronal cultures, we confirmed the concept that PS1 is indispensable for the production of secreted Aβ.

Mutations in presenilin (PS) genes are genetically linked to the early onset of familial Alzheimer's disease (FAD) and are biochemically associated with elevated ratios of Aβ42 to Aβ40, peptides produced by proteolytic cleavages of the βamyloid precursor protein (APP) and deposited as extracellular β-amyloid plaques (1). Mammalian PS consists of two homologous proteins termed PS1 and PS2. They are reported to contain eight transmembrane (TM) domains and a large cytosolic loop region between TM 6 and TM 7 (2, 3). Full-length PS protein undergoes endoproteolytic processing to generate an N-terminal fragment (NTF) and a C-terminal fragment (CTF) of approximately 30 and 20 kDa, respectively (4). Overexpression of exogenous PS does not increase the levels of processed fragments and results in the replacement of endogenous NTF and CTF (5). The cleaved fragments are stable and assemble into a high molecular weight (HMW) complex required for PS proteolytic activity (6).

Overwhelming data have shown that presenilins are required for intramembrane proteolysis of APP and Notch (reviewed in ref. 7), the latter of which plays a critical role in cell fate determination during development. Homozygous PS1-null mice die perinatally because of central nervous system (CNS) and axial skeleton defects that can be largely attributed to a perturbed Notch 1 signaling (8–10). This notion is further supported by reports of Herreman et al. (11) and Donoviel et al. (12) showing that mice lacking both PS1 and PS2 resemble a full Notch 1 null phenotype. However, double-null embryos exhibit additional defects that are not present in either Notch 1 or common Notch downstream target RBPJK mutants, raising the possibility that additional mechanisms independent of Notch may contribute to PS developmental activity (12). This reasoning is corroborated by the fact that PS1 is known to interact with numerous proteins through its hydrophilic loop domain, in particular β-catenin, a molecule essential in Wnt/wingless signaling and cell adhesion (13). PS1 has been shown to down-regulate β-catenin stability and signaling and this activity requires the interaction of the two molecules (14–16). It also associates with the cadherin/catenin complex at adhesive junctions of epithelial cells and at synaptic contacts of the CNS, suggesting a functional role of PS1 in cell–cell interaction (17). Thus, the in vivo activity of the PS1–β-catenin pathway warrants further investigation.

Notch and APP undergo PS-dependent proteolysis. Wolfe et al. (18) reported that presenilins are a class of diaspartyl protease in which the conserved aspartate at residues 257 and 385 (D257 and D385) are the active sites. The conclusion is based on the finding that overexpression of PS1 with the aspartate to alanine mutation at either site (D257A or D385A) results in the replacement of the endogenous PS1 and abolishment of γ-secretase cleavage of APP (18). However, although similar inhibition of Notch activity was observed by the aspartate mutations (19), inconsistent results have been reported with regards to APP processing and Aβ generation in cells expressing the PS1D257A mutation (20–22). One possible explanation for the conflicting observations is the need for endogenous PS1 to be functionally replaced by the mutant PS1, a variable that is difficult to control and to quantify (7). Importantly, these findings have yet to be verified in vivo.

In this study, we wanted to test the effect of PS1 proteolytic activity and PS1–β-catenin association in vivo. Accordingly, we generated transgenic mice expressing human PS1 either with the D257A mutation (hPS1D257A) or deleting amino acids 340–371 (hPS1Δcat), sequences essential to complex with β-catenin, under the transcriptional control of the human Thy-1 (Thy) promoter. By crossing onto PS1-null background, we show that the hPS1D257A fails to rescue the lethality caused by PS1 deficiency and is deficient in APP processing and Aβ production. In contrast, the PS1Δcat transgene is fully active in rescuing the PS1-knockout lethal phenotype and restoring the γ-secretase activity. However, the lack of β-catenin interaction results in the delay of β-catenin degradation and the accumulation of soluble β-catenin in PS1Δcat mutant brain.

Materials and Methods

Antibodies.

The polyclonal PS1-C antibody was raised against amino acids 454 to 467 of PS1. A Glutamate residue was added at the N terminus to enhance the solubility. PSN2, AB14, and 490D and the APP-CTF antibody 369 have been described (15, 23). 4G8 antibody was obtained from Senetik (Maryland Heights, MO). Anti-β-catenin monoclonal antibody was purchased from Sigma. The 9E10 antibody against the c-myc tag was from Santa Cruz Biotechnology.

Vector Construction and Generation of Transgenic and Rescue Mice.

hPS1Δcat and hPS1D257A transgenic vectors were constructed by inserting the human PS1 cDNA deleting amino acids 340–371 (hPS1Δcat) or human PS1 containing the Asp to Ala mutation at residue 257 (hPS1D257A) downstream of the human Thy-1 promoter (24). Transgenic founders were identified by PCR analysis as described (24). Transgenic lines were established by breeding the founders with wild-type B6SJL F1 mice, and their expression levels were determined by Western blot analysis.

Western Blotting and Immunoprecipitation.

Immortalized PS1−/− murine embryonic fibroblast (MEF) cells were transfected with Lipofectamine 2000 Reagent (Invitrogen) by using 2 μg of each DNA according to the manufacturer's recommended conditions. Transfected cells were harvested 24 h later. Notch processing was assayed by Western blot analysis using the anti-c-myc antibody. Notch downstream signaling was measured by using the luciferase reporter system (Promega). Results are expressed as mean ± SD of three independent experiments performed in triplicate.

Western blotting was performed as described (24). For immunoprecipitation, either half adult or whole embryonic brains were lysed at 4°C for 30 min with buffer containing 1.0% digitonin (Sigma), 25 mM Hepes at pH 7.2, 150 mM NaCl, 2 mM DTT, 2 mM EDTA, and 1× proteinase inhibitor. Insoluble material was removed by centrifugation at 17,000 × g for 15 min. The supernatant was preabsorbed with protein A beads followed by incubation with primary antibody and protein A beads overnight at 4°C. The beads were washed five times with lysis buffer and immunoprecipitated proteins were eluted with 2× loading buffer, separated on Tris/Glycine polyacrylamide gels, and subjected to Western blot analysis.

Histology and Staining.

Hematoxylin and eosin staining of mid-sagittal sections of embryonic day 14.5 (E14.5) embryos and adult brains and alcian blue and alizarin red staining of newborn skeletons were performed by using standard procedures as described (8, 24).

Aβ Measurements.

Primary neuronal cultures were prepared from the combined neocortex and hippocampi of E15.5 mice. The brain tissues were dissected under a microscope and the cells were mechanically dissociated, resuspended in Neurobasal medium (Invitrogen), and plated on poly-d-lysine-coated six-well plates at a density of 106 cells per well. Twenty-four hours after plating, the medium was replaced with Neurobasal medium containing 1% B27 supplement (Invitrogen) to promote neuronal survival and discourage nonneuronal cell growth. The cells and conditioned medium were collected for AβX-40 and AβX-42 measurement after 12 days of culturing.

For brain tissues, TBS buffer (4 ml/g of tissue) was used to homogenize brain samples, which were centrifuged at 8,000 × g for 1 h. The pellet was resuspended in 5 M guanidine (twice) and sonicated. Twenty microliters of homogenate was diluted 10-fold with loading buffer. After centrifugation at 8,000 × g for 30 min, the samples were loaded into wells for the detection of Aβ.

Sandwich ELISAs for quantifying mouse and human AβX-40 and AβX-42 were performed as described (25). Briefly, the 96-well immunoassay plates (Nunc) were coated with Aβ40 and Aβ42 end-specific monoclonal antibodies MBC40 and MBC42, respectively. Biotinylated 4G8 (Senetik) was used as detection antibody, which was recognized by streptavidin-conjugated alkaline phosphatase and Attophos fluorescence reporting system (Amersham Pharmacia). Each sample was assayed twice independently for both AβX-40 and AβX-42.

Results

hPS1Δcat and hPS1D257A Exhibit Distinct Developmental Activities.

To determine the function of PS1-β-catenin interaction and the effect of PS1D257A mutation in vivo, we generated multiple lines of transgenic mice expressing human PS1 either with a deletion of amino acids 340–371 of the hydrophilic loop domain (termed hPS1Δcat or Δcat) or containing the D257A mutation (hPS1D257A or DA) driven by the human Thy-1 promoter (Fig. 1) (24). All transgenic lines were viable and overtly normal. Consistent with earlier reports (14, 26), hPS1Δcat protein was endoproteolytically cleaved to generate an NTF indistinguishable from that of wild-type hPS1 and a truncated CTF (Fig. 1 B and D). In contrast, hPS1D257A formed only a full-length protein (FL, Fig. 1C), in agreement with previous observations that this mutation does not undergo endoproteolysis. Regardless of the expression levels of the transgene, endogenous PS1-CTF remained constant; indicating that expression of transgene did not replace the endogenous fragments in vivo. Similar results were also observed when wild-type PS1 or PS1 containing familial Alzheimer's disease mutation (PS1A246E) was expressed, reflecting the physiological expression of the transgenes by the human Thy-1 promoter (24).

Figure 1.

Figure 1

Open in a new tab

Biochemical characterization of hPS1Δcat and hPS1D257A transgenic and rescue mice. (A) Schematic representation of the transgenic constructs. hPS1Δcat is human PS1 cDNA deleting amino acids 340–371 of the hydrophilic loop sequence. hPS1D257A contains an aspartate to alanine mutation at residue 257 of PS1. Each construct was inserted into the pHZ024 vector, which includes the human Thy-1 (Thy) promoter and SV40 small t and poly(A) sequences (24). (B) Brain expression profiles of hPS1Δcat transgenic mice assayed by Western blot analysis using the human PS1-specific NTF antibody PSN2. Δcat-3 and Δcat-6 represents two hPS1Δcat transgenic lines: wt, wild-type nontransgenic mouse brain; and 17-3, a previously identified human wild-type PS1 transgenic line (24). Hybridization with an anti-β-actin was used as loading control. (C) Brain expression profiles of hPS1D257A (DA) transgenic mice (lines 3, 4, 6, and 7) by using the PSN2 antibody, which detected only full-length (FL) hPS1 in transgenic lines (Upper). The same brain lysates were also hybridized with the PS1 loop antibody 490D to detect PS1-CTF (Lower). wt, wild-type nontransgenic mouse brain. (D) Western blot analysis of hPS1Δcat (lines 3 and 6) and hPS1D257A (lines 3, 4, 6, and 7) transgenes expressed on mouse PS1-null (rescue or R) background. E14.5 embryonic brains were hybridized with either the PSN2 antibody (Upper) or the PS1-C antibody, which was raised against the C-terminus end of hPS1 (Lower). Both the wild-type hPS1 rescue (17-3R) and hPS1Δcat rescue (Δcat-3R and Δcat-6R) brains formed indistinguishable PS1-NTF, and normal or truncated CTF, respectively. However, the hPS1D257A rescue lines (DA-3R, DA-4R, DA-6R, and DA-7R) formed only FL protein. (E) Interaction of β-catenin by hPS1Δcat and hPS1D257A. Brain samples of DA-7R, Δcat-3R, or 17-3R were immunoprecipitated with an anti-β-catenin antibody and Western blotted with the PS1 antibody AB14 and the anti-β-catenin antibody. PS1Δcat failed to associate with β-catenin.

To determine the developmental activities of hPS1Δcat and hPS1D257A, the transgenic lines were crossed with PS1+/− mice to generate offspring lacking endogenous PS1 but containing the human PS1 transgenes (termed Δcat-R or DA-R, where R represents “rescue”). Western blot analysis of the rescue brains by using antibodies against the PS1 N terminus (PSN2) and C terminus (PS1C) confirmed that hPS1Δcat protein was normally cleaved to generate an NTF and a truncated CTF (Fig. 1D). In contrast, no endoproteolysis could be identified in multiple lines of hPS1D257A rescue mutants (Fig. 1D). To determine whether hPS1Δcat and hPS1D257A were able to associate with β-catenin, brain samples of hPS1Δcat (line 3), hPS1D257A (line 7), and wild-type human PS1 (17-3) rescue mice were immunoprecipitated with an anti-β-catenin antibody followed by Western blotting with a PS1 antibody. PS1-NTF and PS1-FL proteins could be readily recovered from β-catenin immunoprecipitates in wild-type hPS1 and hPS1D257A rescue brains respectively, demonstrating that both wild-type PS1 and PS1D257A mutant complex with β-catenin (Fig. 1E). However, no PS1 could be detected from β-catenin immunoprecipitates in hPS1Δcat rescue brains (Fig. 1E), confirming that hPS1Δcat is defective in PS1–β-catenin interaction. Reverse experiments using PS1 immunoprecipitation followed by β-catenin Western blot analysis yielded the same result (data not shown).

Histological examinations of sectioned embryos at E14.5 of development and whole-mount skeleton staining of newborn pups revealed that hPS1Δcat rescue mice exhibited normal brain structure whereas the skeleton defect of the PS1 null was either partially (line 6) or completely (line 3) corrected (Fig. 2 and Table 1). The degree of rescue correlates with embryonic expression of the transgene and is roughly comparable to that of wild-type hPS1 rescue (24). In sharp contrast, multiple lines (lines 3, 4, 6, and 7) of hPS1D257A rescue embryos invariably exhibited developmental defects indistinguishable from those of PS1-null, including CNS hemorrhage and axial skeleton patterning defect (Fig. 2 and Table 1). These results demonstrate that hPS1Δcat possesses normal PS1 developmental activity, whereas hPS1D257A is completely negative.

Figure 2.

Figure 2

Open in a new tab

Histology and skeleton staining of hPS1Δcat and hPS1D257A rescue mice. (A) Hematoxylin and eosin staining of mid-sagittal sections of E14.5 embryos, showing CNS hemorrhage and axial skeleton defects in both PS1-knockout (PS1−/−) and hPS1D257A rescue (PS1−/−; hPS1D257A) mice, whereas the hPS1Δcat rescue (PS1−/−; hPS1Δcat) mice looked normal. (B) Alcian blue and alizarin red staining of newborn skeletons revealing similar findings. WT, wild-type control.

Table 1.

Summary of rescue phenotypes by hPS1Δcat and hPS1D257A transgenes

Transgene Line Expression* Rescue activity†
Lethality Skeleton Hemorrhage
hPS1‡ 17-3 ++++ Yes Complete Yes
hPS1Δcat 3 +++ Yes Complete Yes
6 ++ Yes Partial Yes
hPS1D257A 3 + No No No
4 ++ No No No
6 + No No No
7 +++ No No No
Open in a new tab
*

Expression levels were assayed by Western blot analysis of PS1 in the brains of E14.5 embryos and adult mice, assuming that the antibody recognizes the full-length and cleaved products with equal affinity. 

†

Rescue activity was determined based on morphological examinations of hematoxylin and eosin staining of sagittal sections of E12.5–E17.5 embryos and whole-mount skeleton staining of newborn pups, using PS1-knockout mice as baseline. 

‡

The wild-type hPS1 rescue line 17-3 was used as a positive control. 

hPS1D257A Is Inactive in Notch Processing and Downstream Signaling.

We reported earlier that hPS1Δcat exhibits normal Notch processing activity in vitro (15), which correlates with its developmental function in vivo. Studies have shown that PS1 aspartate mutants are impaired in the proteolytic cleavages of Notch and variably in APP processing (18, 20, 27). However, the effects of the D257A mutation on downstream Notch signaling and Aβ expression have been controversial. One explanation for the variable results is the presence of endogenous PS1 in virtually all of the studies. Therefore, the mutant phenotype derived from D257A expression depends on the ability to functionally “replace” endogenous presenilin. To directly determine the effect of the D257A mutation, we first used PS1-deficient fibroblast cells and cotransfected hPS1D257A, and a truncated Notch-1 construct NΔE. PS1-dependent proteolytic activity was determined by measuring the production of Notch intracellular domain (NICD) (28). As expected, expression of hPS1D257A did not result in detectable NICD production whereas transfection of wild-type hPS1 led to the efficient cleavage of NΔE to NICD (Fig. 3A). We next measured the Notch-mediated downstream signaling by cotransfecting hPS1D257A with the CBF1 (C-promoter binding factor 1 or RBPJk)-luciferase reporter construct and by quantifying the luciferase activities (29). In complete agreement with the failure of hPS1D257A to process Notch, this mutant was defective in activating the Notch downstream signaling mediated by the CBF1 promoter (Fig. 3B). These results establish that proteolytic processing of Notch is a prerequisite for its signaling activity.

Figure 3.

Figure 3

Open in a new tab

Notch processing and downstream signaling by hPS1D257A mutant. (A) PS1−/− fibroblast cells were cotransfected with the constitutively active Notch construct NΔE plus pcDNA3 empty vector (vector), pcDNA3-hPS1D257A (D257A), or pcDNA3-wild-type hPS1 (hPS1), respectively. PS1 proteolytic activity was measured by the appearance of Notch intracellular domain (NICD) by Western blotting. Transfection with NICD was used as size control. (B) Notch-mediated downstream signaling was determined by cotransfecting the PS1−/− cells with the CBF-luciferase reporter plus empty vector (bar 1), hPS1D257A (bar 2), or wild-type hPS1 (bar 3), respectively, followed by quantifying the luciferase activity. AU, arbitrary unit.

hPS1D257A Is Defective in APP Cleavage and Production of Aβ.

Having documented that hPS1D257A is inactive in processing and signaling Notch, we next asked whether the mutation affects APP proteolysis and, most importantly, Aβ production. Embryonic and adult brains were isolated from PS1−/−; hPS1D257A (lines 6 and 7) and PS1−/−; hPS1Δcat (line 3) rescue mice, respectively, and the levels of full-length APP (APP-FL) and APP-CTF were assayed by using the APP C-terminal antibody 369 (23). As expected, levels of APP were low in fetal brains as compared with adult brains (lanes 1–4 vs. 5–7; Fig. 4A). APP-CTF, representing products from both α- and β-secretase activities, were barely detectable in PS1+/+ controls but were substantially higher in PS1−/− brains (Fig. 4A, lanes 3 and 4, respectively). APP-CTF was also elevated in hPS1D257A rescue brains similar to that of PS1-null (lanes 1 and 2, Fig. 4A). As predicted, no appreciable difference could be seen in the amount of APP CTF in PS1−/− mice rescued with wild-type or Δcat hPS1 (Fig. 4A, lanes 5, 6, and 7), indicating that the expression of hPS1Δcat restores the γ-secretase processing of APP.

Figure 4.

Figure 4

Open in a new tab

APP processing and Aβ production by hPS1D257A and hPS1Δcat transgenes. (A) Western blot analysis of APP-FL and APP-CTF levels in E14.5 (lines 1–4) and adult (lanes 5–7) brains. DA-6R and DA-7R: PS1−/−; hPS1D257A (lines 6 and 7), Δcat-3R: PS1−/−; hPS1Δcat, (line 3), and 17-3: wild-type hPS1. (B and C) Endogenous AβX-40 (B) and AβX-42 (C) levels in conditioned medium of primary neuronal cultures of PS1−/− (n = 8), PS1−/−; hPS1D257A (line 7) (PS1−/−; DA-7, n = 10), and PS1+/+ (n = 10) brains. (D and E) Endogenous AβX-40 (D) and AβX-42 (E) levels in the brains of control (PS1+/+, n = 8), hPS1Δcat (line 3) rescue (PS1−/−; Δcat-3, n = 8), and wild-type hPS1 rescue (PS1−/−; 17-3, n = 8) mice. Levels represent mean ± standard deviation.

We next examined levels of AβX-40 and AβX-42 peptides by using a highly sensitive and specific sandwich ELISA method (25). This assay allows detection of multiple PS-dependent γ-secretase cleavage products, including those of BACE1 at both site 1 and site 11 of Aβ (30). Conditioned medium was collected from primary neuronal cultures of E15.5 embryonic brains, and the levels of secreted AβX-40 and AβX-42, were quantified. Here we show that, consistent with the vast majority of the analysis and in disagreement with Armogida et al. (31), both endogenously secreted mouse AβX-40 and AβX-42 were greatly reduced in PS1−/− neuronal cultures compared with that of PS1+/+ controls (Fig. 4 B and C). Furthermore, in contrast to reports that PS1D257A lead to the normal production of Aβ (20–22), our results showed that secreted AβX-40 and AβX-42 were dramatically decreased in PS1−/−; hPS1D257A rescue cultures to levels not significantly different from that of PS1-null cells (Fig. 4 B and C). Therefore, the aspartate-257 is essential for PS1-mediated APP processing at the γ-secretase site and the latter is necessary for secreted Aβ expression.

Consistent with the normal production of APP-CTF by hPS1Δcat, measurement of Aβ from adult brains showed that the expression of wild-type hPS1 or hPS1Δcat on PS1−/− background restored Aβ levels to similar degrees (Fig. 4 D and E). It is interesting to note that although AβX-42 was completely recovered, the expression of either hPS1 transgene only partially restored AβX-40 levels (Fig. 4D). This result is in line with our earlier analysis suggesting a possible differential regulation of Aβ40 and Aβ42 production between mouse and human PS1 (24).

Enhanced β-Catenin Stability in hPS1Δcat Rescue Mice.

Both hPS1Δcat rescue lines (3 and 6) were viable and fertile. Similar to wild-type hPS1 rescue (16), these mice develop epidermal hyperplasia, a phenotype associated with the loss of PS1 in skin (data not shown). As both PS1 and β-catenin are expressed and form a complex in the brain and hPS1Δcat mutant fails to associate with β-catenin, we assessed the effect PS1–β-catenin interaction on soluble β-catenin levels and its stability in the CNS. Neuronal cultures were derived from hPS1Δcat and wild-type hPS1 rescue embryos, and levels of soluble β-catenin and its stability were determined. Western blot analysis showed that there was approximately a 2-fold increase in the steady-state levels of soluble β-catenin in hPS1Δcat rescue cells as compared with wild-type hPS1 rescue cells (Fig. 5A). Consistent with its ability to interact with β-catenin, hPS1D257A effectively regulated β-catenin to levels similar to that of wild-type hPS1 (Fig. 5A). Elevated β-catenin in hPS1Δcat rescue cells is accompanied by a delayed ubiquitination (Fig. 5B) and turnover (Fig. 5C) of this β-catenin pool, as assayed by treating the cells with the proteasome inhibitor MG132 and the protein synthesis inhibitor cycloheximide, followed by Western blotting, respectively. The functional consequence of the elevated β-catenin in hPS1Δcat rescue mice requires further investigation. Preliminary immunostaining analysis with antibodies against glial fibrillary acidic protein (GFAP) and synaptophysin failed to detect significant abnormalities in hPS1Δcat rescue brains (not shown).

Figure 5.

Figure 5

Open in a new tab

Soluble β-catenin levels and stability in hPS1Δcat rescue mutant. (A) Higher steady-state levels of soluble β-catenin in PS1−/−; hPS1Δcat (line 3) (Δcat-3R) neurons as compared with that of wild-type hPS1 rescue (17-3R) samples. In contrast, no appreciable differences in soluble β-catenin levels were detected between PS1−/−; hPS1D257A (line 7) (DA-7R) and wild-type hPS1 rescue (17-3R). (B) Delayed β-catenin ubiquitination by PS1Δcat. Neuronal cultures of PS1−/−; 17-3 or PS1−/−; Δcat-3 mice were treated with the proteasome inhibitor MG132 (20 μM) for 0, 30, or 60 min followed by immunoblotting with the anti-β-catenin antibody to visualize nonubiquinated (β-catenin) and multiubiquitinated β-catenin (Ub-β-catenin). (C) Impaired β-catenin turnover by PS1Δcat. PS1−/−; 17-3 and PS1−/−; Δcat-3 cultures were treated with cycloheximide (25 mg/ml) for 0, 15, 30, or 60 min. Soluble β-catenin was extracted and subjected to immunoblotting with the anti-β-catenin antibody.

Discussion

Effect of PS1-β-Catenin Interaction on PS1 Developmental Activity.

PS1–β-catenin interaction has been implicated in the regulation of cytosolic β-catenin and its downstream signaling as well as cadherin/catenin complex stability (15, 16, 32). Consistent with our previous data from PS1-null fibroblasts and keratinocytes (15, 16), here we show that PS1 is a negative regulator of β-catenin in the brain and that this function requires the interaction of the two molecules. The lack of PS1-β-catenin association in hPS1Δcat rescue mice leads to delayed β-catenin turnover and higher steady-state levels of the protein in CNS. However, the fact that hPS1Δcat is able to rescue the neuronal and embryonic defects of the PS1-null mice demonstrates that this activity is dispensable for PS1 developmental function. Although β-catenin resides in the PS1 high molecular weight complex required for its proteolytic activity, it is not an essential component, because as disruption of PS1–β-catenin association preserves PS1-mediated proteolysis. Our result is in line with the in vitro finding reported by Saura et al. (26). Although no overt CNS phenotype can detected, it remains to be tested whether the hPS1Δcat rescue mice develop CNS or other abnormalities at an older age.

Activity of hPS1D257A on Development and Aβ Production.

Most of the studies investigating PS proteolytic activities use nonneuronal cell culture systems and many rely on the “replacement” phenomenon in which the endogenous presenilins have to be replaced by overexpressing foreign PS derivatives, presumably by competing for limited cellular factors. However, it is difficult to control the degree of replacement and it is not known whether the high levels of overexpression of the transfected protein, which is required to replace the endogenous PS, may lead to other nonspecific adverse effects. Differences in the degree of replacement may be the reason for the inconsistent and often conflicting data reported (7). The role of PS in Notch and APP processing is further complicated by a recent report showing that, although Notch proteolysis is blocked, endogenous Aβ are normally produced in PS-deficient fibroblasts (31). This observation challenges the vast majority of published data demonstrating that PS activity is a prerequisite for Aβ production (33–35). Because of the central role of presenilins in Alzheimer's disease pathogenesis, this fundamental issue needs to be further investigated.

Our analysis offers the following advantages: (i) Endogenously secreted mouse AβX-40 and AβX-42 are measured, thus avoiding the potential distortion of the enzyme complex/substrate ratio because of overexpression of APP; (ii) the effect of the D257A mutation is determined by introducing the mutant allele onto PS1-null background in vivo, which eliminates the need to replace endogenous presenilin by overexpressing the mutant protein; and (iii) AβX-40 and AβX-42 secreted from primary neuronal cultures rather than from transformed nonneuronal cell lines are measured.

Several positive correlations can be derived by analyzing the effects of hPS1D257A mutant expressed on PS1-null background. First, we show that PS1D257A is completely defective in Notch processing and downstream signaling in vitro and developmental activity in vivo. Therefore, the in vitro Notch assays correlate with the in vivo developmental outcomes of PS1 and its derivatives. However, in light of the emerging numbers of presenilin proteolytic substrates identified such as ErbB-4 and CD44 (36, 37), it is likely that the D257A mutation would affect these targets and associated pathways as well. Accordingly, the additional phenotypes of the presenilin double-null embryos could be attributed by defects of these presenilin proteolytic targets. Therefore, the in vitro Notch assays should be considered as a rapid method to assess the presenilin proteolytic activity, the latter of which mediates its developmental function.

Second, we report that hPS1D257A leads to the concurrent accumulation of APP-CTF and the reduction of secreted Aβ peptides in neuronal cultures. Therefore, the aspartate-257 is required for PS1 to process APP at the γ-secretase site to generate Aβ. This result is in line with the original finding of Wolfe et al. (18) but different from several follow-up studies (20–22) documenting that PS1D257A is capable of generating Aβ regardless of the clear accumulation of APP-CTF. Although the reason for these inconsistent findings is yet to be settled, a noticeable difference of our system is that, in contrast to the above studies where the mutant protein is overexpressed resulting in the replacement of endogenous PS1, the human Thy-1 promoter leads to physiological expression of the transgenes without replacing the endogenous protein (Fig. 1C and ref. 24). Although it can be argued that the failure for the hPS1D257A to exert any effect on PS1 developmental function and γ-secretase activity is because of its relatively low levels of expression, similar expression of both wild-type hPS1 and hPS1Δcat could restore both activities when expressed on PS1-null background (Table 1 and ref. 24). Therefore, the negative rescue outcome of hPS1D257A is most likely the result of the mutation per se rather than expression levels.

In summary, we took advantage of our established PS1 “rescue” system to assess the effects PS1–β-catenin association and PS1D257A mutation on PS1 developmental activity, APP processing, and Aβ production in an in vivo physiological context. Several important conclusions can be made from the current study: (i) PS1 in Notch and β-catenin signaling pathways can be genetically and functionally uncoupled; (ii) PS1 is indispensable for secreted Aβ peptide production in neurons; and (iii) aspartate-257 of PS1 is critical for Notch and APP proteolysis and subsequent Notch signaling and Aβ generation, respectively, thus providing strong support that these two pathways are mediated through the same mechanisms.

Acknowledgments

We thank David Trinh and Tahira Zaidi for expert technical assistance, Y. Wu for the initial phase of the project, R. Wang for help in neuronal culturing, and X. Wu for advice. We are grateful to H. Mori for the gift of PSN2 antibody, H. Xu for providing 369 and AB14 antibodies, and R. Kopan for the Notch constructs. This work was supported by Grants from the National Institutes of Health (NS40039), Alzheimer's Association (11RG-00-2221), and American Health and Assistance Foundation (A2000019). H.Z. is a New Scholar of the Ellison Medical Foundation.

Abbreviations

PS1

presenilin 1

APP

β-amyloid precursor protein

TM

transmembrane

Aβ

β-amyloid peptides

NTF

N-terminal fragment

CTF

C-terminal fragment

CNS

central nervous system

NICD

Notch intracellular domain

En

embryonic day n

Footnotes

This paper was submitted directly (Track II) to the PNAS office.

References

  • 1.Selkoe D J. Trends Cell Biol. 1998;8:447–453. doi: 10.1016/s0962-8924(98)01363-4. [DOI] [PubMed] [Google Scholar]
  • 2.Doan A, Thinakaran G, Borchelt D R, Slunt H H, Ratovitsky T, Podlisny M, Selkoe D J, Seeger M, Gandy S E, Price D L, Sisodia S S. Neuron. 1996;17:1023–1030. doi: 10.1016/s0896-6273(00)80232-9. [DOI] [PubMed] [Google Scholar]
  • 3.Li X, Greenwald I. Proc Natl Acad Sci USA. 1998;95:7109–7114. doi: 10.1073/pnas.95.12.7109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Thinakaran G, Borchelt D R, Lee M K, Slunt H H, Spitzer L, Kim G, Ratovitsky T, Davenport F, Nordstedt C, Seeger M, et al. Neuron. 1996;17:181–190. doi: 10.1016/s0896-6273(00)80291-3. [DOI] [PubMed] [Google Scholar]
  • 5.Thinakaran G, Harris C L, Ratovitsky T, Davenport F, Slunt H H, Price D L, Borchelt D R, Sisodia S S. J Biol Chem. 1997;272:28415–28422. doi: 10.1074/jbc.272.45.28415. [DOI] [PubMed] [Google Scholar]
  • 6.Yu G, Chen F, Levesque G, Nishimura M, Zhang D M, Levesque L, Rogaeva E, Xu D, Liang Y, Duthie M, et al. J Biol Chem. 1998;273:16470–16475. doi: 10.1074/jbc.273.26.16470. [DOI] [PubMed] [Google Scholar]
  • 7.Strooper B D, Annaert W. Nat Cell Biol. 2001;3:E221–E225. doi: 10.1038/ncb1001-e221. [DOI] [PubMed] [Google Scholar]
  • 8.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. Nature (London) 1997;387:288–292. doi: 10.1038/387288a0. [DOI] [PubMed] [Google Scholar]
  • 9.Shen J, Bronson R T, Chen D F, Xia W, Selkoe D J, Tonegawa S. Cell. 1997;89:629–639. doi: 10.1016/s0092-8674(00)80244-5. [DOI] [PubMed] [Google Scholar]
  • 10.Handler M, Yang X, Shen J. Development (Cambridge, UK) 2000;127:2593–2606. doi: 10.1242/dev.127.12.2593. [DOI] [PubMed] [Google Scholar]
  • 11.Herreman A, Hartmann D, Annaert W, Saftig P, Craessaerts K, Serneels L, Umans L, Schrijvers V, Checler F, Vanderstichele H, et al. Proc Natl Acad Sci USA. 1999;96:11872–11877. doi: 10.1073/pnas.96.21.11872. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Donoviel D B, Hadjantonakis A-K, Ikeda M, Zheng H, St. George Hyslop P, Bernstein A. Genes Dev. 1999;13:2801–2810. doi: 10.1101/gad.13.21.2801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Peifer M, Polakis P. Science. 2000;287:1606–1609. doi: 10.1126/science.287.5458.1606. [DOI] [PubMed] [Google Scholar]
  • 14.Kang D E, Soriano S, Frosch M P, Collins T, Naruse S, Sisodia S S, Leibowitz G, Levine F, Koo E H. J Neurosci. 1999;19:4229–4237. doi: 10.1523/JNEUROSCI.19-11-04229.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Soriano S, Kang D E, Fu M, Pestell R, Chevallier N, Zheng H, Koo E H. J Cell Biol. 2001;152:785–794. doi: 10.1083/jcb.152.4.785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Xia X, Qian S, Soriano S, Wu Y, Fletcher A M, Wang X-J, Koo E H, Wu X, Zheng H. Proc Natl Acad Sci USA. 2001;98:10863–10868. doi: 10.1073/pnas.191284198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Georgakopoulos A, Marambaud P, Efthimiopoulos S, Shioi J, Cui W, Li H C, Schutte M, Gordon R, Holstein G R, Martinelli G, et al. Mol Cell. 1999;4:893–902. doi: 10.1016/s1097-2765(00)80219-1. [DOI] [PubMed] [Google Scholar]
  • 18.Wolfe M S, Xia W, Ostaszewski B L, Diehl T S, Kimberly W T, Selkoe D J. Nature (London) 1999;398:513–517. doi: 10.1038/19077. [DOI] [PubMed] [Google Scholar]
  • 19.Ray W J, Yao M, Mumm J, Schroeter E H, Saftig P, Wolfe M, Selkoe D J, Kopan R, Goate A M. J Biol Chem. 1999;274:36801–36807. doi: 10.1074/jbc.274.51.36801. [DOI] [PubMed] [Google Scholar]
  • 20.Capell A, Steiner H, Romig H, Keck S, Baader M, Grim M G, Baumeister R, Haass C. Nat Cell Biol. 2000;2:205–211. doi: 10.1038/35008626. [DOI] [PubMed] [Google Scholar]
  • 21.Morihara T, Katayama T, Sato N, Yoneda T, Manabe T, Hitomi J, Abe H, Imaizumi K, Tohyama M. Brain Res Mol Brain Res. 2000;85:85–90. doi: 10.1016/s0169-328x(00)00229-1. [DOI] [PubMed] [Google Scholar]
  • 22.Kim S H, Leem J Y, Lah J J, Slunt H H, Levey A I, Thinakaran G, Sisodia S S. J Biol Chem. 2001;276:43343–43350. doi: 10.1074/jbc.M108245200. [DOI] [PubMed] [Google Scholar]
  • 23.Buxbaum J D, Gandy S E, Cicchetti P, Ehrlich M E, Czernik A J, Fracasso R P, Ramabhadran T V, Unterbeck A J, Greengard P. Proc Natl Acad Sci USA. 1990;87:6003–6006. doi: 10.1073/pnas.87.15.6003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Qian S, Jiang P, Guan X M, Singh G, Trumbauer M E, Yu H, Chen H Y, Van de Ploeg L H, Zheng H. Neuron. 1998;20:611–617. doi: 10.1016/s0896-6273(00)80999-x. [DOI] [PubMed] [Google Scholar]
  • 25.Sun X, Sato S, Murayama O, Murayama M, Park J-M, Yamaguchi H, Takashima A. Neurosci Lett. 2002;321:61–64. doi: 10.1016/s0304-3940(01)02583-6. [DOI] [PubMed] [Google Scholar]
  • 26.Saura C A, Tomita T, Soriano S, Takahashi M, Leem J Y, Honda T, Koo E H, Iwatsubo T, Thinakaran G. J Biol Chem. 2000;275:17136–17142. doi: 10.1074/jbc.M909624199. [DOI] [PubMed] [Google Scholar]
  • 27.Berezovska O, Jack C, McLean P, Aster J C, Hicks C, Xia W, Wolfe M S, Kimberly W T, Weinmaster G, Selkoe D J, Hyman B T. J Neurochem. 2000;75:583–593. doi: 10.1046/j.1471-4159.2000.0750583.x. [DOI] [PubMed] [Google Scholar]
  • 28.Kopan R, Schroeter E H, Weintraub H, Nye J S. Proc Natl Acad Sci USA. 1996;93:1683–1688. doi: 10.1073/pnas.93.4.1683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Lu F M, Lux S E. Proc Natl Acad Sci USA. 1996;93:5663–5667. doi: 10.1073/pnas.93.11.5663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Cai H, Wang Y, McCarthy D, Wen H, Borchelt D R, Price D L, Wong P C. Nat Neurosci. 2001;4:233–234. doi: 10.1038/85064. [DOI] [PubMed] [Google Scholar]
  • 31.Armogida M, Petit A, Vincent B, Scarzello S, Alves da Costa A, Checler F. Nat Cell Biol. 2001;3:1030–1033. doi: 10.1038/ncb1101-1030. [DOI] [PubMed] [Google Scholar]
  • 32.Baki L, Marambaud P, Efthimiopoulos S, Georgakopoulos A, Wen P, Cui W, Shioi J, Koo E, Ozawa M, Friedrich V L, Jr, Robakis N K. Proc Natl Acad Sci USA. 2001;98:2381–2386. doi: 10.1073/pnas.041603398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.De Strooper B, Saftig P, Craessaerts K, Vanderstichele H, Guhde G, Annaert W, Von Figura K, Van Leuven F. Nature (London) 1998;391:387–390. doi: 10.1038/34910. [DOI] [PubMed] [Google Scholar]
  • 34.Herreman A, Serneels L, Annaert W, Collen D, Schoonjans L, De Strooper B. Nat Cell Biol. 2000;2:461–462. doi: 10.1038/35017105. [DOI] [PubMed] [Google Scholar]
  • 35.Zhang Z, Nadeau P, Song W, Donoviel D, Yuan M, Bernstein A, Yankner B A. Nat Cell Biol. 2000;2:463–465. doi: 10.1038/35017108. [DOI] [PubMed] [Google Scholar]
  • 36.Ni C-Y, Murphy M P, Golde T E, Carpenter G. Science. 2001;294:2179–2181. doi: 10.1126/science.1065412. [DOI] [PubMed] [Google Scholar]
  • 37.Okamoto I, Kawano Y, Murakami D, Sasayama T, Araki N, Miki T, Wong A J, Saya H. J Cell Biol. 2001;155:755–762. doi: 10.1083/jcb.200108159. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES