J Biol Chem, Vol. 273, Issue 51, 33909-33914, December 18, 1998

Abrogation of the Presenilin 1/-Catenin Interaction and Preservation of the Heterodimeric Presenilin 1 Complex following Caspase Activation^*

Giuseppina Tesco, Tae-Wan Kim^§, Anke Diehlmann^¶, Konrad Beyreuther^¶, and Rudolph E. Tanzi

From the Genetics and Aging Unit, Massachusetts General Hospital and Harvard Medical School, Charlestown, Massachusetts 02129 and the ^¶ Zentrum für Molekulare Biologie Heidelberg, INF 282, 69120 Heidelberg, Federal Republic of Germany

	ABSTRACT

Top Abstract Introduction Procedures Results & Discussion References

-Catenin has previously been shown to interact with presenilin 1 (PS1) in transfected cells. Here we report that -catenin co-immunoprecipitates with the endogenous C-terminal fragment of presenilin 1 (PS1-CTF) but not with the endogenous CTF of presenilin 2 (PS2-CTF) in H4 human neuroglioma cells. During staurosporine (STS)-induced cell death, -catenin and PS1-CTF undergo a caspase-mediated cleavage. After 12 h of STS treatment, the -catenin·PS1-CTF interaction is abrogated. While PS1-CTF immunoprecipitated with all caspase-cleaved species of -catenin, -catenin holoprotein did not co-immunoprecipitate with the "alternative" caspase-derived PS1-CTF (PS1-aCTF). Thus, the abrogation of the -catenin·PS1-CTF complex was due to caspase cleavage of PS1-CTF. -Catenin co-immunoprecipitated with PS1-NTF, but only when PS1-NTF was associated with PS1-CTF. Even though PS1-NTF·CTF complex stability was not altered by caspase cleavage, its ability to bind -catenin was abolished. Thus, while the PS1-NTF·CTF complex is preserved after caspase cleavage, it may no longer be fully functional.

	INTRODUCTION

Top Abstract Introduction Procedures Results & Discussion References

Missense mutations in presenilin 1 (PS1)¹ and presenilin 2 (PS2) are responsible for roughly 30-40% of early onset familial Alzheimer's disease (1). The presenilins are transmembrane proteins localized in the endoplasmic reticulum (ER) and Golgi (2). To date, over 50 different PS1 mutations have been described, and 3 FAD mutations have been identified in PS2. Two main clusters of mutations in exon 5 and exon 8 are observed in PS1. In native cell lines and brain, full-length PS1 is not detectable as the protein constitutively undergoes endoproteolytic cleavage resulting in the production of a fragment of ~20 kDa (C-terminal fragment or PS1-CTF) and a fragment of ~30 kDa (N-terminal fragment or PS1-NTF) (3-5). It has recently been reported that the NTF and CTF of both presenilins are associated in heterodimeric complexes in cultured cells (6, 7) and brain (8) and that -catenin is also present in this complex (9). Because FAD mutations in these proteins are associated with an increase in A-(1-42) both in vivo and in vitro (10-13), it has been suggested that PS1 and PS2 could play a role in the processing of amyloid precursor protein in the ER and/or Golgi. Another series of studies reported that PS1 and PS2 are highly homologous with Sel-12, a protein that facilitates the Notch pathway in Caenorhabditis elegans (14, 15).

To further elucidate the biological function of presenilins, efforts have been made to identify molecules that interact with presenilins. A recent study has shown that PS1 interacts with -catenin, a homologue of -catenin. In addition, -catenin has been shown to bind full-length PS1 in human embryonic kidney 293 cells overexpressing PS1 (16).

-Catenin is a multifunctional protein that plays a key role in the Wingless pathway and in the regulation of cell-cell adhesion (for review see Ref. 17). -Catenin is able to bind directly to E-cadherins and connects the adherens junctions to the actin cytoskeleton via interaction with -catenin. The cytoplasmic level of -catenin is highly regulated. -Catenin is degraded by the proteasome following phosphorylation. In the steady state, glycogen synthase kinase 3 phosphorylates -catenin at its N terminus. When the Wingless pathway is activated, glycogen synthase kinase 3 is inhibited, resulting in an increase in the cytoplasmic level of -catenin and making the protein available for transportation to the nucleus where it can bind the T-cell factor/lymphoid enhancer-binding factor family of transcription factors. -Catenin degradation is also activated by various negative regulators of the Wingless pathway. For example, the adenomatous polyposis coli gene product binds and down-regulates -catenin.

Recently it has been shown that during apoptosis, -catenin undergoes caspase-mediated cleavage (18). The execution phase of apoptosis involves the cleavage and activation of a cascade of caspases that are responsible for the cleavage of various "cell death" substrates (for review see Ref. 19). Caspase cleavage of -catenin occurs at the N- and C-terminal regions of the protein resulting in the disruption of -catenin·-catenin binding. Since -catenin binds actin, caspase cleavage of -catenin affects the organization of actin filaments and contributes to dismantling actin anchorage of cells during apoptosis (18). During apoptosis, the PS1-CTF and PS2-CTF also undergo caspase-mediated "alternative" cleavage resulting in the production of smaller fragments (PS1-aCTF and PS2-aCTF) (20, 21). The presence of FAD missense mutations in overexpressed PS2 has recently been shown to lead to increased production of PS2-aCTF (20).

Since a previous report showed that -catenin could be co-immunoprecipitated with PS1 in transfected cells, one of the aims of this study was to determine whether -catenin also interacts with endogenous PS1-CTF in native human neuroglioma H4 cells. Moreover, we set out to determine whether, during apoptosis, caspase-mediated cleavage affects the interaction between PS1 and -catenin and complex formation between PS1-NTF and CTF.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results & Discussion References

Antibodies-- We used two different anti--catenin monoclonal antibodies identified as anti--catenin_C-100 raised against the last 100 aa of the C terminus of -catenin (Zymed Laboratories Inc.) and anti--catenin_571-781 raised against aa 571-781 of the C terminus of -catenin (Transduction Laboratories). Anti--catenin_571-781 is also available as horseradish peroxidase-conjugated antibody (anti--catenin HRPO, Transduction Laboratories). For immunodetection of PS1-CTF, we used 3 different antibodies: PS1Loop, a polyclonal antibody raised against the large hydrophilic loop region of PS1 (gift from Dr. G. Thinakaran and Dr. S. Sisodia (3)); 4627, polyclonal antibody raised against aa 457-467 of PS1 C terminus (gift from Dr. D. Selkoe (5)), and APS 18, a monoclonal antibody raised against aa 313-334 of the loop region of PS1 (7). For immunodetection of PS2-CTF, we used 2 different antibodies: PS2Loop, a polyclonal antibody raised against the large hydrophilic loop region of PS2 (gift from Dr. G. Thinakaran and Dr. S. Sisodia (3)), and APS 26, a monoclonal antibody raised against aa 317-334 of the loop region of PS2. For immunodetection of PS1-NTF we used Ab14 antibody, a polyclonal antibody raised against 1-25 aa of PS1 (gift of Drs M. Seeger and S. Gandy (3)). A list of all antibodies used in this study is provided in Table I.

View this table: [in this window] [in a new window]   Table I -Catenin, PS1, and PS2 antibodies

Cell Culture, Induction, and Inhibition of Cell Death-- H4 human neuroglioma cells were cultured in DMEM (high glucose) containing 10% heat-inactivated fetal calf serum, 100 units/ml penicillin, 100 µg/ml streptomycin, 2 mM L-glutamine. For the induction of apoptosis, cells were seeded at a density of 5 × 10⁶ cells for each 150-mm Petri dish. After 48 h, the cell layer was washed twice with phosphate-buffered saline, and the standard medium was replaced with serum-free DMEM containing 1 µM staurosporine (STS) (Calbiochem). A time course experiment was performed: H4 cells were collected at time 0, 4, 6, 8, 12, 16, and 24 of STS treatment. Cells were pelleted by centrifugation washed once with phosphate-buffered saline. For apoptosis inhibition cells were incubated with 100 µM zVAD-FMK (Enzyme System Products) and 1 µM STS or STS only for 16 h in serum-free DMEM and collected as before. Cells were treated with zVAD for 1 h before STS addition. The characterization of apoptotic process in naive H4 cells in response to STS treatment including various parameters such as cell death measurement by trypan blue exclusion, caspase 3 activation, and PARP cleavage has been shown previously (20).

Western Blot Analysis and Immunoprecipitation-- Cell pellets from each time point following treatment with STS were detergent-extracted on ice using IP buffer (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 2 mM EDTA, 0.5% Nonidet P-40) plus protease inhibitors. The lysates were centrifuged at 1000 × g for 10 min at 4 °C. The supernatants were collected, and total proteins were quantitated by the BCA protein assay kit (Pierce). For Western blot analysis, 100 µg of total protein of each sample were subjected to SDS-polyacrylamide gel electrophoresis using 4-20% gradient Tris/glycine gels under reducing conditions (Novex). Proteins were transferred to polyvinylidene difluoride membrane (Bio-Rad) using a semi-dry electrotransfer system (Hoefer). The blots were blocked with 5% non-fat dry milk in TBST (25 mM Tris, pH 7.4, 137 mM NaCl, 0.15% Tween 20) for 1.5 h, incubated with primary antibodies (anti--catenin:horseradish peroxidase-conjugated (HRPO), 1:1000 in 5% non-fat dry milk TBST; PS1 Loop, 1:2500 in TBST) for 1 and 1.5 h, respectively and incubated with secondary antibodies (horseradish peroxidase-conjugated anti-mouse or anti-rabbit antibodies, 1:5000 (Pierce)) in 5% non-fat dry milk TBST for 1 h. Between steps, the blots were washed with TBST for 30 min. The blots were visualized using the ECL or ECL plus Western blot detection system (Amersham). For immunoprecipitation (IP), 2 mg of total proteins for each sample were used. For quantitative co-IP experiments, increasing amounts of total protein were used (0.375 mg, 0.750 mg, 1.5 mg, 3 mg). For co-immunoprecipitation of PS1-NTF and PS1-CTF cell pellets from time point 0 and 24 of STS treatment were detergent-extracted on ice using IP buffer (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 2 mM EDTA, 1% digitonin). The samples were precleared either with protein A (15 µl) or goat anti-mouse-IgG magnetic bead-conjugated (45 µl) (Perceptive Diagnostics) overnight in the cold room. For immunoprecipitation with monoclonal antibodies goat anti-mouse IgG magnetic bead-conjugated were precoated using 2 µg of anti--catenin_C-100 (Zymed Laboratories Inc.), 1.5 µg of anti--catenin_571-781 (Transduction Laboratories), 4 µg of APS18, 4 µg of APS26 for each sample. After preclearing, the magnetic beads were collected using a magnetic bead collector and then discarded; the precleared lysates were collected and incubated with anti-mouse-IgG magnetic beads-conjugated precoated with the monoclonal antibodies listed above, or with 1 µl of PS1 Loop or PS2 Loop polyclonal antibody or Ab14 antibody, and incubated overnight in the cold room. The samples immunoprecipitated with polyclonal antibodies were further incubated with Protein A magnetic beads-conjugated for 2 h in the cold room. Immunoprecipitates were washed twice in IP buffer then collected using a magnetic bead collector, heated at 50 °C for 10 min in sample buffer, and subjected to SDS-polyacrylamide gel electrophoresis and Western blotting using anti--catenin HRPO, 1:1000, or PS1 Loop, 1:2500, PS2 Loop, 1:2500, or 4627 antibody, 1:2500, Ab14 antibody, 1:2500.

	RESULTS AND DISCUSSION

Top Abstract Introduction Procedures Results & Discussion References

-Catenin and PS1-CTF Binding in H4 Human Neuroglioma Cells-- In this study, we set out to examine whether endogenous forms of PS1-CTF interact with -catenin in H4 human neuroglioma cells and whether caspase-mediated cleavage affects this interaction. Detergent lysates prepared from H4 cells were subjected to immunoprecipitation (IP) using two different monoclonal antibodies: anti--catenin_C-100 and anti--catenin_571-781 (Fig. 1A, lanes 1 and 2, respectively). Western blot analysis performed with anti--catenin HRPO antibody detected -catenin in both immunoprecipitates (Fig. 1A, upper panel, lanes 1 and 2) and in the straight lysate (lane 3). When the same blot was immunostained with PS1 Loop antiserum, endogenous PS1-CTF was detected as a doublet of ~20 kDa and ~23 kDa, as previously reported in H4 cells (20) (Fig. 1A, lower panel, lanes 1 and 2). Thus, both -catenin antibodies were able to co-IP the PS1-CTF. To confirm the binding between -catenin and PS1-CTF, we carried out the reverse experiment using two different antibodies raised against different epitopes of the large hydrophilic loop of PS1: a monoclonal antibody, APS 18, and a polyclonal antiserum, PS1 Loop. PS1-CTF was immunodetected in both immunoprecipitates (Fig. 1A, lower panel, lanes 4 and 5) and in the straight lysate (lane 6). The presence of -catenin in PS1 immunoprecipitates was shown by immunostaining the same blot with anti--catenin antibody (Fig. 1A, upper panel, lanes 4 and 5).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   A, co-immunoprecipitation of -catenin and the PS1-CTF. H4 cell lysate was immunoprecipitated with the indicated antibodies. The precipitates and the straight lysate (lanes 3 and 6) were immunoblotted with anti--catenin antibody HRPO (Transduction Laboratories) (upper panel) and PS1 Loop (lower panel). Note that anti--catenin antibody detects -catenin after immunoprecipitation with both APS18 and PS1 Loop (upper panel) and that PS1 Loop detects PS1-CTF after immunoprecipitation with both anti--cateninC-100 and anti--catenin571-781 antibodies. B, lack of co-immunoprecipitation of PS2-CTF and -catenin. H4 cell lysate was immunoprecipitated with antibody APS26, or PS2 Loop antibody, (lanes 1 and 2). Immunodetection of the precipitates and the straight lysate (lane 3) using anti--catenin HRPO antibody is reported in the upper panel. Note that PS2-CTF was not co-immunoprecipitated with -catenin, which is detected only in straight lysate (lane 3). Western blot analysis of the same precipitates performed with PS2 Loop antibody detected PS2-CTF in all lanes (lower panel). C, quantitative co-immunoprecipitation of -catenin and PS1-CTF. Increasing amounts of total proteins were immunoprecipitated using anti--catenin antibody. The precipitates were immunodetected with anti--catenin antibody HRPO (upper panel) and PS1 Loop antibody (lower panel). Note that as increasing amounts of -catenin are immunoprecipitated from increasing amounts of total protein, increasing amounts of the PS1-CTF are also co-immunoprecipitated.

To demonstrate the specificity of the interaction between -catenin and PS1-CTF, we performed the same co-IP experiment using several antibodies as negative controls. Cell lysates were immunoprecipitated with two different antibodies raised against the large hydrophilic loop of PS2, ASP 26 (monoclonal) and PS2 Loop (polyclonal) and with anti--tubulin and anti-low density lipoprotein receptor related protein (LRP) antibodies. Although PS2-CTF was successfully immunoprecipitated by both PS2 antibodies as shown in the Western blot with PS2 Loop antibody (Fig. 1B, lower panel, lanes 1 and 2), -catenin was not immunodetected in the immunoprecipitates (Fig. 1B, upper panel, lanes 1 and 2), but was present in the straight lysate (Fig. 1B, upper panel, lane 3). In negative control experiments, LRP and -tubulin antibodies did not co-immunoprecipitate with -catenin (data not shown). To further test the specificity of the binding between -catenin and PS1-CTF, we performed a quantitative co-IP experiment (Fig. 1C). As increasing amounts of total proteins were immunoprecipitated with anti--catenin antibody, increasing amounts of -catenin were detected when immunoprecipitates were subjected to Western blot analysis using anti--catenin antibody (Fig. 3, upper panel). Developing the same blot with PS1 Loop antibody, we were able to show that increasing amounts of PS1-CTF co-immunoprecipitated with correspondingly increasing amounts of -catenin (Fig. 3C, lower panel). Taken together, these data demonstrate a specific interaction between endogenous -catenin and endogenous PS1-CTF in H4 cells.

Effect of Apoptosis on -Catenin and PS1 Cleavage-- Since it has been previously reported that both -catenin (18) and the presenilins (20) are substrates for caspases, we investigated the effects of apoptosis on the processing of -catenin and PS1 in our cellular model. Apoptosis was induced by treating H4 cells with staurosporine, and time course experiments were performed. When Western blot analysis of cell lysates from different time points was carried out with anti--catenin_C-100 antibody, -catenin was not detected after 6 h of STS treatment (data not shown). We next reprobed the blot using anti--catenin_571-781 antibody and detected fragments of different sizes (~90 kDa, ~65 kDa, and intermediate species; Fig. 2, upper panel, time points 6-24 h). Thus, during the progression of STS-induced cell death, -catenin holoprotein undergoes sequential cleavage by caspases at multiple sites, resulting in the appearance of an ~90-kDa species at the 6-h time point, because of the removal of a fragment of ~2 kDa from the C terminus of -catenin. The 90-kDa species is then further cleaved into a smaller fragment of ~65 kDa that becomes the prevalent species in the latest time points (12-24 h), as described previously (26). Thus, during apoptosis, -catenin is sequentially cleaved in H4 cells in the same type of pattern that was described previously for NIH3T3 fibroblasts and Madin-Darbin canine kidney cells. When the same blot was probed with PS1 Loop antibody, PS1-CTF and PS1-aCTF were immunodetected (Fig. 2, lower panel). PS1-aCTF was only detected after 8 h of STS treatment. During the course of STS-induced cell death, the amount of PS1-aCTF increased (time points 8-24) while PS1-CTF, which serves as a caspase substrate, was decreased in amount. While some of the light extra bands were nonspecific, the slightly higher molecular mass bands have previously been identified as phosphorylated PS1-CTF by protein kinase C (6). As expected, staurosporine, a broad spectrum protein kinase inhibitor, which was used to induce apoptosis in these cells, decreased and even eliminated these bands during the time course, as it also inhibits protein kinase C. These data show that both -catenin and PS1-CTF are substrates for caspases and exhibit different temporal patterns of caspase-mediated cleavage. These data also suggest the involvement of different species of caspase, given that the PS1 caspase site does not contain the caspase 3 consensus sequence (DXXD) (in contrast to the caspase cleavage sites of -catenin).

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Analysis of -catenin and PS1 cleavage during apoptosis in H4 cells. H4 cells were treated with 1 µM staurosporine. Cells were harvested and lysed in lysis buffer at the time points indicated. Upper panel, Western blot analysis of -catenin using anti--catenin antibody HRPO. Note that after 6 h of treatment, -catenin full-length (FL) was cleaved into ~90- and ~65-kDa fragments (indicated by arrows). The ~65-kDa fragment becomes the predominant species during later points in the time course. Lower panel, Western blot analysis of the PS1-CTF using PS1 Loop antibody. Note that the PS1-aCTF can be detected after 8 h and increases in amount during the time course coinciding with the increased appearance of the ~65-kDa cleavage product of -catenin.

Effect of Caspase Cleavage on -Catenin·PS1 Binding-- Since both -catenin and PS1 are cleaved during apoptosis, we next investigated the effects of caspase-mediated cleavage on -catenin·PS1-CTF binding. For this purpose, we performed co-IP experiments from cell lysates corresponding to each time point of the STS time course experiments, using anti--catenin_571-781 antibody (Fig. 3, A and C), APS18, and PS1 Loop (Fig. 3, B and D, respectively). When cell lysates were immunoprecipitated with anti--catenin_571-781 antibody, PS1-CTF was immunodetected by both PS1 Loop (Fig. 3A, lower panel) and by 4627 antibody (Fig. 3C, lower panel) at time points 0, 6, and 8. Neither PS1-CTF nor PS1-aCTF co-immunoprecipitated with -catenin at time points 12, 16, or 24. Immunodetection of -catenin in the same blot (Fig. 3, A and C, upper panel) showed that all caspase-cleaved species were immunoprecipitated in the same temporal pattern that we previously observed in Western blot analyses (Fig. 2, upper panel). To further confirm that -catenin·PS1-CTF binding is abolished after 12 h of STS treatment, we performed complementary co-IP experiments using two different antibodies raised against the large hydrophilic loop of PS1 in the same set of experiments. When PS1-CTF was immunoprecipitated using APS18, all species of -catenin were immunodetected at time points 0, 6, and 8 but not at 12, 16, and 24 (Fig. 3B, upper panel). Western blot analysis performed using PS1 Loop detected only PS1-CTF in the immunoprecipitates (Fig. 3B, lower panel). Since APS18 did not immunoprecipitate PS1-aCTF, we used cell lysates from the same time points to immunoprecipitate with PS1 Loop antibody (Fig. 3D). PS1 Loop antibody was able to pull down both PS1-CTF and PS1-aCTF as detected by the 4627 antibody (Fig. 3D, lower panel). Again, -catenin co-immunoprecipitated with the PS1-CTF, but only at time points 0, 6, and 8 (as observed when the same blot was developed with anti--catenin HRPO antibody, Fig. 3D, upper panel).

View larger version (51K): [in this window] [in a new window]   Fig. 3.   Co-immunoprecipitation of -catenin and the PS1-CTF during apoptosis in H4 cells. A and B, lysates from each time point were immunoprecipitated with the indicated antibodies. Immunoprecipitates were analyzed by Western blot analysis using anti--catenin antibody HRPO (upper panel) and PS1 Loop antibody (lower panel). Note that -catenin co-immunoprecipitated with PS1-CTF but not PS1-aCTF. At the time points in which PS1-aCTF becomes predominant, the binding between -catenin and PS1 was abolished (A, lower panel). In panel B, APS18, which recognizes only PS1-CTF and not PS1-aCTF (in the lower panel the PS1-CTF is detected by PS1 Loop antibody), was able to co-immunoprecipitate all of the different caspase-cleaved species of -catenin (~90 and ~65 kDa and intermediate fragments, indicated by arrows), but only during the first 8 h of STS treatment. After 8 h of STS treatment the binding between -catenin and PS1-CTF was abolished. C and D, lysates from each time point were immunoprecipitated with the indicated antibodies. Immunoprecipitates were detected with anti--catenin HRPO antibody (upper panel) or with 4627 (polyclonal antibody raised against PS1 C terminus 457-467 aa) (lower panel). Note that PS1 Loop was able to immunoprecipitate the PS1-CTF and the PS1-aCTF (D, lower panel) which, in turn, co-immunoprecipitated with -catenin in the same pattern as that observed with APS18 (B, upper panel). The anti--catenin antibody was able to immunoprecipitate all caspase-cleaved species of -catenin (C, upper panel) which co-immunoprecipitated with PS1-CTF in a pattern similar to that observed in panel A (although detected with 4627 antibody).

These data show that the binding between -catenin and PS1-CTF was not detected in lysates from cells treated with STS for 12 h or longer. This suggests that caspase-mediated cleavage occurs at the binding site of one or both proteins and abrogates the interaction of these proteins. Since -catenin holoprotein (Fig. 3, A and B, lower panels, time 0) and all -catenin caspase-cleaved species, including the ~65-kDa fragment of -catenin, which became the predominant species in the later time points, were able to co-immunoprecipitate with PS1-CTF (Fig. 3, A and B, lower panels, time 6 and 8) and vice versa (Fig. 3, B and D, upper panels, same time points), it is unlikely that the -catenin binding site is lost at the later time points. Conversely, no binding was observed at time points 12, 16, and 24, when the amount of PS1-CTF decreased and the amount of PS1-aCTF increased. At the same time points, PS1-CTF was still detectable (Fig. 3, B and D, lower panel) but probably may have been insufficiently abundant to immunoprecipitate and detect -catenin. In addition, PS1-aCTF did not co-immunoprecipitate with any of the -catenin caspase-cleaved species.

If the abrogation of -catenin·PS1-CTF complex was due to the caspase-mediated cleavage of the -catenin binding site of PS1-CTF and not to the caspase-mediated cleavage of -catenin, -catenin holoprotein, which was not present in the later time points, should have co-immunoprecipitated with PS1-CTF but not with PS1-aCTF. To test this hypothesis, we immunoprecipitated -catenin in a sample containing lysates from both time point 0 and time point 24. Thus, the mixed sample contained -catenin holoprotein, PS1-CTF (from time point 0 lysate, Fig. 4A, lane 1), -catenin caspase-cleaved species, and PS1-aCTF (from time point 24 lysate, Fig. 4A, lane 2). As expected, -catenin holoprotein co-immunoprecipitated with PS1-CTF at time point 0 (Fig. 4A, lane 3) and at time point 24, the prevalent ~65-kDa -catenin fragment did not co-immunoprecipitate with either PS1-CTF or PS1-aCTF (Fig. 4A, lane 4). In the mixed sample, -catenin holoprotein co-immunoprecipitated with PS1-CTF, but not with PS1-aCTF (Fig. 4 A, lane 5). These data show that the abrogation of the -catenin·PS1 binding after 24 h of STS treatment (Fig. 4A, lane 4) was due to caspase-mediated cleavage of PS1-CTF.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   A, caspase-mediated cleavage of PS1-CTF abrogates -catenin·PS1-CTF interaction. H4 cells treated with 1 µM STS for 24 h (24) or not treated (0) were harvested and incubated in IP buffer containing 0.5% Nonidet P-40. Lysates from time points 0, 24, and 0 + 24 (a mixed sample containing both lysates from time point 0 and time point 24) were immunoprecipitated with anti--catenin571-781 antibody. Western blot analysis of straight lysates from time points 0 and 24 (lanes 1 and 2) and of immunoprecipitates from time points 0, 24, and 0 + 24 (lanes 3, 4, and 5) was performed using anti--catenin HRPO antibody (upper panel) and PS1 Loop antibody (lower panel). B, co-immunoprecipitation of PS1-NTF with -catenin in normal cells and lack of binding in apoptotic cells. Lysates of H4 cells from time points 0 and 24 following STS treatment were prepared in IP buffer containing 1% digitonin and immunoprecipitated with Ab14 antibody. Western blot analysis of immunoprecipitates from time points 0 and 24 (lanes 1 and 2) was performed using anti--catenin HRPO antibody (upper panel), Ab14 antibody (middle panel), and PS1 Loop antibody (lower panel). C, effect of zVAD treatment on abrogation of binding between -catenin and the PS1-CTF by caspase-mediated cleavage. H4 cells were incubated with 1 µM STS in serum-free DMEM with or without 100 µM zVAD and harvested after 16 h of incubation. Cells were treated with zVAD for 1 h before STS addition. Western blot analysis of cell lysates using both anti--catenin antibody (upper panel) and PS1 Loop (lower panel) showed that zVAD treatment (lane 1) was able to inhibit the caspase-mediated cleavage of both -catenin and PS1-CTF induced by STS treatment (lane 2) and preserve -catenin·PS1 binding (lane 3).

Since it has previously been shown that PS1-NTF and PS1-CTF form an oligomeric complex (6-8), which also contains -catenin (9), we next investigated whether -catenin could be co-immunoprecipitated with PS1-NTF. PS1-NTF was not detected in immunoprecipitates from lysates extracted with 0.5% Nonidet P-40 (Fig. 4A, middle panel). Therefore, we first determined appropriate detergent conditions for co-immunoprecipitating PS1-NTF and PS1-CTF. For this purpose, we used as a detergent 1%, 0.5%, 0.25% Nonidet P-40, 1% digitonin, or 0.25% dodecyl-D-maltoside. Immunoprecipitation was carried out using Ab14, and the reverse experiment was performed using APS18 for immunoprecipitation. The optimal detergent conditions for co-immunoprecipitating PS1-NTF with PS1-CTF and PS1-NTF with -catenin was 1% digitonin (data not shown). H4 cells from time points 0 and 24 of STS treatment were incubated in lysis buffer containing 1% digitonin, and the lysates were immunoprecipitated with Ab14 (Fig. 4B, lanes 1 and 2, respectively). Immunoprecipitates were then subjected to Western blot analysis using PS1 Loop antibody (Fig. 4B, lower panel). PS1-NTF co-immunoprecipitated with PS1-CTF at time 0 (Fig. 4B, lower panel, lane 1), and PS1-NTF was also able to bind both PS1-CTF and PS1-aCTF at time point 24 (Fig. 4B, lower panel, lane 2). When the same blot was immunostained with anti--catenin HRPO antibody, -catenin was co-immunoprecipitated by Ab14 at time 0, but the binding was lost at time point 24, even though PS1-NTF could be successfully immunoprecipitated in both samples (Fig. 4B, middle panel, lanes 1 and 2). PS1-NTF co-immunoprecipitated with -catenin only in detergent conditions that allowed PS1-NTF·PS1-CTF binding. In addition, when PS1-CTF was cleaved by caspases, not only was the binding between -catenin and PS1-CTF abolished, the -catenin·PS1-NTF binding was also lost. Moreover, binding of -catenin to PS1-CTF did not require the presence of PS1-NTF. Collectively these data show that -catenin binds to PS1-NTF but only via its interaction with PS1-CTF. It should also be noted that caspase-mediated cleavage of PS1-CTF did not interfere with the stability of the PS1-NTF·PS1-CTF complex. In fact, PS1-NTF was still able to interact with PS1-aCTF.

To test whether the disruption of the -catenin·PS1 interaction was actually caused by caspase cleavage, we blocked caspase activation using a broad spectrum caspase inhibitor (zVAD-FMK) and performed co-IP experiments. Protein extracts from cells treated with STS in the presence or absence of zVAD for 16 h (Fig. 4C, lanes 1 and 2, respectively) were subjected to Western blot analysis and immunodetected either with anti--catenin HRPO antibody (Fig. 4C, upper panel) or PS1 Loop. zVAD treatment was able to completely inhibit both -catenin and PS1 cleavage (Fig. 4C, lanes 1 and 2). The same lysates were immunoprecipitated with anti--catenin antibody, and immunoprecipitates were detected with PS1 Loop. PS1-CTF co-immunoprecipitated with -catenin in zVAD plus STS-treated cell extract (Fig. 4C, lower panel, lane 3), but not in the cells treated with STS alone (Fig. 4C, lower panel, lane 4), although -catenin species were present in both samples as demonstrated by Western blot analysis using an anti--catenin antibody (Fig. 4C, upper panel). Thus, inhibition of caspase-mediated cleavage of PS1 and -catenin by zVAD treatment preserves the -catenin·PS1-CTF interaction. Collectively, these data indicate that -catenin binds the normal PS1-CTF in H4 cells, while caspase-mediated cleavage abolishes this interaction.

The initiation of the cell death program includes the proteolytic activation of caspases which, through autoactivation and activation of other caspases, results in the cleavage of a large set of cell death substrates (19). Caspase-mediated cleavage events promote the "execution" stage of apoptosis by disrupting several types of interactions. These include DNA·transcription factor-DNA (e.g. SAF-A, Sp1; Refs. 22 and 23), RNA·RNA-binding protein (e.g. mdm2; Ref. 24), and protein·protein (e.g. -catenin·-catenin binding, focal adhesion kinase·paxillin; Refs. 18 and 25) interactions. Since -catenin has been shown to play a role in the Wingless pathway and in cadherin-mediated cell-to-cell adhesion, diverse physiological functions of -catenin may be mediated by different molecular interactions. It has previously been reported that caspase-mediated cleavage of -catenin abrogates its interaction with -catenin, thereby disrupting actin organization in the apoptotic cell (18). Alterations in the actin cytoskeleton lead to the loss of cell-cell adhesion and changes in cell shape that are associated with anchorage-related apoptosis or anoikis (26). We now show that the -catenin·PS1 interaction is also abrogated by caspase activation. It is possible that the abrogation of -catenin·PS1 binding might also contribute to the cytoskeleton alterations that lead to cellular changes that are essential for the execution of apoptosis. Moreover, since it has been shown that during apoptosis there is an increase in the production of PS1-aCTF in cells expressing FAD-linked mutant forms of PS1,² it is possible that in Alzheimer's patients with PS1 mutations the -catenin·PS1-CTF interaction could be adversely affected by the increased production of PS1-aCTF, thereby leading to cellular changes that make neurons more susceptible to programmed cell death. Interestingly, in contrast, caspase cleavage did not abolish the PS1-NTF·CTF complex. However, following caspase cleavage, the PS1-NTF·CTF complex can no longer bind -catenin. Thus, while the PS1-NTF·CTF complex is preserved after caspase cleavage, it may no longer be fully functional. Future studies will be necessary to explore how caspase-mediated effects on presenilin interactions may contribute to AD pathogenesis.

	ACKNOWLEDGEMENTS

We thank Drs. S. Gandy and M. Seeger for the generous gift of the Ab14 antibody, Dr. D. Selkoe for the 4627 antibody, and Drs. S. Sisodia and G. Thinakaran for the PS1 Loop and PS2 Loop antibodies.

	FOOTNOTES

^* These studies were supported by grants from the NINDS and NIA, National Institutes of Health, and the Alzheimer's Association Temple Award (to R. E. T.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of an Extendicare Foundation/The John Douglas French Alzheimer's Foundation fellowship.

^§ Recipient of the Partners Investigator (Nesson) Award.

To whom correspondence should be addressed: Genetics and Aging Unit, Dept. of Neurology, Massachusetts General Hospital-East, 149 13th Street, Charlestown, MA 02129. Tel.: 617-726-6845; Fax: 617-726-5677; E-mail: tanzi{at}helix.mgh.harvard.edu.

The abbreviations used are: PS1, presenilin 1; PS2, presenilin 2; PS1-CTF, C-terminal fragment of PS1; PS1-NTF, N-terminal fragment of PS1; PS1-aCTF, alternative cleaved PS1 C-terminal fragment; ER, endoplasmic reticulum; FAD, familial Alzheimer's disease; STS, staurosporine; IP, immunoprecipitation; aa, amino acid; DMEM, Dulbecco's modified Eagle's medium; HRPO, horseradish peroxidase; PARP, poly(ADP-ribose)polymerase; zVAD, z-Val-Ala-Asp-fluoromethylketone.

² D. M. Kovacs, R. Mancini, J. Henderson, S. Na, T-W. Kim, and R. E. Tanzi, submitted for publication.

	REFERENCES

Top Abstract Introduction Procedures Results & Discussion References

1. Tanzi, R. E. (1998) in Scientific American Molecular Neurology (Martin, J. B., ed), pp. 55-75, Scientific American, New York
2. Kovacs, D. M., Fausett, H. J., Page, K. J., Kim, T.-W., Mori, 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]
3. Thinakaran, G., Borchelt, D., Lee, M., Slunt, H., Spitzer, L., Kim, G., Ratovitsky, T., Davenport, F., Nordstedt, C., Seeger, M., Hardy, J., Levey, A. I., Gandy, S. E., Jenkins, N. A., Copeland, N. G., Price, D. L., and Sisodia, S. S. (1996) Neuron 17, 181-190[CrossRef][Medline] [Order article via Infotrieve]
4. Kim, T.-W., Pettingell, W. H., Hallmark, O. G., Moir, R. D., Wasco, W., and Tanzi, R. E. (1997) J. Biol. Chem. 272, 11006-11010[Abstract/Free Full Text]
5. 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. 3, 325-337[CrossRef][Medline] [Order article via Infotrieve]
6. Seeger, M., Nordstedt, C., Petanceka, S., Kovacs, D. M., Gouras, G. K., Hahne, S., Fraser, P., Levesque, L., Czernik, A. J., St. George-Hyslop, P., Sisodia, S. S., Thinakaran, G., Tanzi, R. E., Greengard, P., and Gandy, S. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 5090-5094[Abstract/Free Full Text]
7. Capell, A., Grünberg, J., Pesold, B., Diehlmann, A., Citron, M., Nixon, R., Beyreuther, K., Selkoe, D. J., and Haass, C. (1998) J. Biol. Chem. 273, 3205-3211[Abstract/Free Full Text]
8. Thinakaran, G., Regard, J. B., Bouton, C. M. L., Hariis, C. L., Price, D. L., Borchelt, D. R., and Sisodia, S. S. (1998) Neurobiol. Dis. 4, 438-453[CrossRef][Medline] [Order article via Infotrieve]
9. Yu, G., Chen, F., Levesque, G., Nishimura, M., Zhang, D. M., Levesque, L., Rogaeva, E., Xu, D., Liang, Y., Duthie, M., St George-Hyslop, P. H., and Fraser, P. E. (1998) J. Biol. Chem. 273, 16470-16475[Abstract/Free Full Text]
10. Scheuner, D., Eckman, C., Jensen, M., Song, X., Citron, M., Suzuki, N., Bird, T. D., Hardy, J., Hutton, M., Kukull, W., Larson, E., Levy-Lahad, E., Viitanen, M., Peskind, E., Poorkaj, P., Schellenberg, G., Tanzi, R., Wasco, W., Lannfelt, L., Selkoe, D., and Younkin, S. (1996) Nat. Med. 2, 864-870[CrossRef][Medline] [Order article via Infotrieve]
11. Citron, M., Westaway, D., Xia, W., Carlson, G., Diehl, T., Levesque, G., Johnson-Wood, K., Lee, M., Seubert, P., Davis, A., Kholodenko, D., Motter, R., Sherrington, R., Perry, B., Yao, H., Strome, R., Lieberburg, I., Rommens, J., Kim, S., Schenk, D., Fraser, P., St George Hyslop, P., and Selkoe, D. J. (1996) Nat. Med. 3, 67-72
12. Duff, K., Eckman, C., Zehr, C., Yu, X., Prada, C. M., Perez-tur, J., Hutton, M., Buee, L., Harigaya, Y., Yager, D., Morgan, D., Gordon, M. N., Holcomb, L., Refolo, L., Zenk, B., Hardy, J., and Younkin, S. (1996) Nature 383, 710-713[CrossRef][Medline] [Order article via Infotrieve]
13. Borchelt, D. R., Thinakaran, G., Eckman, C. B., Lee, M. K., Davenport, F., Ratovitsky, T., et al.. (1996) Neuron 17, 1005-1013[CrossRef][Medline] [Order article via Infotrieve]
14. Levitan, D., and Greenwald, I. (1995) Nature 377, 351-354[CrossRef][Medline] [Order article via Infotrieve]
15. Levitan, D., Doyle, T. G., Brousseau, D., Lee, M. K., Thinakaran, G., Slunt, H. H., Sisodia, S. S., and Greenwald, I. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 14940-14944[Abstract/Free Full Text]
16. Zhou, J., Liyanage, U., Medina, M., Ho, C., Simmons, A. D., Lovett, M., and Kosik, K. S. (1997) NeuroReport 8, 1489-1494[Medline] [Order article via Infotrieve]
17. Barth, A. I., Näthke, I. S., and Nelson, W. J. (1997) Curr. Opin. Cell Biol. 9, 683-690[CrossRef][Medline] [Order article via Infotrieve]
18. Brancolini, C., Lazarevic, D., Rodriguez, J., and Schneider, C. (1997) J. Cell Biol. 139, 759-771[Abstract/Free Full Text]
19. Villa, P., Kaufmann, S. H., and Earnshaw, W. C. (1997) Trends Biochem. Sci. 22, 388-393[CrossRef][Medline] [Order article via Infotrieve]
20. Kim, T.-W., Pettingell, W. H., Jung, Y.-K., Kovacs, D. M., and Tanzi, R. E. (1997) Science 277, 373-376[Abstract/Free Full Text]
21. Kim, T.-W., and Tanzi, R. E. (1997) Curr. Opin. Neurobiol. 7, 683-688[CrossRef][Medline] [Order article via Infotrieve]
22. Gohring, F., Schwab, B. L., Nicotera, P., Leist, M., and Fackelmayer, F. O. (1997) EMBO J. 16, 7361-7371[CrossRef][Medline] [Order article via Infotrieve]
23. Piedrafita, F. J., and Pfahl, M. (1997) Mol. Cell. Biol. 17, 6348-6358[Abstract]
24. Chen, L., Marechal, V., Moreau, J., Levine, A. J., and Chen, J. (1997) J. Biol. Chem. 272, 22966-22973[Abstract/Free Full Text]
25. Levkau, B., Herren, B., Koyama, H., Ross, R., and Raines, E. W. (1998) J. Exp. Med. 187, 579-586[Abstract/Free Full Text]
26. Frisch, S. M., and Ruoslahti, E. (1997) Curr. Opin. Cell Biol. 7, 701-706