| Human Molecular Genetics | Pages |
©1999 Oxford University Press |
Aberrant splicing in the presenilin-1 intron 4 mutation causes presenile Alzheimer's disease by increased A[beta]42 secretion
Introduction
Results
Genetic analysis
cDNA analysis
Site-directed mutagenesis
cDNA transfections
In vivo analysis of PSEN1 expression
A[beta] secretion
Discussion
Materials And Methods
Mutation analysis
Polymorphism analysis
cDNA analysis
Sequence analysis
PSEN1 cDNA constructs
In vitro transcription/translation
Transfection
cDNA synthesis
Immunoblot analysis
Immunoprecipitation
Immunofluorescence microscopy
A[beta] measurement
Acknowledgements
References
Aberrant splicing in the presenilin-1 intron 4 mutation causes presenile Alzheimer's disease by increased A[beta]42 secretion
Received April 23, 1999; Revised and Accepted June 8, 1999
We previously described a splice donor site mutation in intron 4 of presenilin-1 (PSEN1) in two patients with autopsy-confirmed early-onset Alzheimer's disease (AD). Here we provide evidence that the intron 4 mutation is present in four additional unrelated early-onset AD cases, that the mutation segregates in an autosomal dominant manner and that all cases have one common ancestor. We demonstrate that the intron 4 mutation produces three different transcripts, two deletion transcripts ([Delta]4 and [Delta]4cryptic) and one insertion transcript (insTAC), by aberrant splicing. The deletion transcripts result in the formation of C-truncated (~7 kDa) PSEN1 proteins while the insertion transcript produces a full-length PSEN1 with one extra amino acid (Thr) inserted between codons 113 and 114 (PSEN1 T113-114ins). The truncated proteins were not detectable in vivo in brain homogenates or lymphoblast lysates of mutation carriers. In vitro HEK-293 cells overexpressing [Delta]4, [Delta]4cryptic or insTAC PSEN1 cDNAs showed increased A[beta]42 secretion (~3.4 times) only for the insertion cDNA construct. Increased A[beta]42 production was also observed in brain homogenates. Our data indicate that in the case of intron 4 mutation, the AD pathophysiology results from the presence of the PSEN1 T113-114ins protein comparable with cases carrying dominant PSEN1 missense mutations.
INTRODUCTION
Alzheimer's disease (AD) is a devastating neurodegenerative disorder and the major cause of senile dementia. Mutations in three genes, the amyloid precursor protein (APP) (1), the presenilin-1 (PSEN1) (2) and the presenilin-2 (PSEN2) genes (3,4), are involved in the etiology of early-onset AD. The majority of the mutations were found in PSEN1 on chromosome 14 (5). The PSEN1 gene comprises 13 exons, of which exons 3-12 code for a protein of 467 amino acids (6-8). Different PSEN1 transcripts were reported resulting from alternative splicing of a 12 bp sequence in exon 3, coding for the amino acids Val Arg Ser Gln (VRSQ) (6,7), and of exon 8 (2).
PSEN1 encodes a serpentine-like protein, located in the endoplasmic reticulum and Golgi complex (9-11). The PSEN1 protein has six to eight transmembrane (TM) domains with the N-terminus, the large sixth hydrophilic loop (HL-VI) and the C-terminus protruding into the cytoplasm (9,10,12). PSEN1 is endoproteolytically processed within the region of HL-VI encoded by exon 9 (13), resulting in the production of an ~28 kDa N-terminal fragment (NTF) and ~18 kDa C-terminal fragment (CTF). The endoproteolytic processing is a tightly regulated mechanism in which CTF and NTF accumulate to a saturable level in a 1:1 stoichiometry (13). The exact biological function of PSEN1 remains unknown. Recently, it was suggested that PSEN1 is either directly or indirectly involved in the processing of APP into A[beta]42 (14-17). This finding corroborates previous observations that mutations in PSEN1 probably cause AD by a gain-of-function mechanism. PSEN1 mutations result in increased A[beta]42 secretion in plasma and fibroblasts of mutation carriers (18), in media of transfected cells and in brains of transgenic mice (18-24). A[beta]42 is the 42 amino acid form of A[beta] and is believed to be pathogenic since it is more prone to aggregation, accelerating its deposition in brain (25). PSEN1 null mice are embryonic lethal and have decreased A[beta]42 formation (26-28). PSEN1 mutants rescue the embryonic lethality of PSEN1 null mice with an increase in A[beta]42 secretion (29,30). Altered APP metabolism with increased A[beta]42 formation is therefore considered an intrinsic property of mutant PSEN1 proteins.
With the exception of the in-frame deletion of exon 9 ([Delta]9) (31-34), all PSEN1 mutations reported are missense mutations (5). However, we recently reported a single nucleotide (G) deletion in the consensus sequence of the splice donor site of intron 4 in two autopsy-confirmed cases with early-onset AD (35). We demonstrated that this mutation results in abnormally spliced PSEN1 transcripts lacking the entire exon 4 ([Delta]4) or part of exon 4 ([Delta]4cryptic), due to exon skipping or cryptic splicing within exon 4, respectively (Fig. 1A-C). In both deletion transcripts the reading frame is shifted, resulting in a premature stop codon (Fig. 1A-C). Both [Delta]4 and [Delta]4cryptic transcripts with and without the VRSQ were detected in both AD patients, although at much lower abundance than wild-type (WT) PSEN1 mRNA (35). Based on the truncation site and the location of the premature stop codons, we predicted the formation of a C-truncated PSEN1 protein (+VRSQ) of 70 amino acids for [Delta]4 and 77 for [Delta]4cryptic, corresponding to a peptide of ~7 kDa (Fig. 1A-C). In the absence of knowledge of the biology of the truncated PSEN1 transcripts and/or proteins, we hypothesized that this truncating mutation leads to AD pathology through haploinsufficiency of full-length PSEN1 (35).
Figure 1. Proteins generated by WT PSEN1 (A), PSEN1 [Delta]4 (B), PSEN1 [Delta]4cryptic (C) and insTAC (D) transcripts. A 1 bp deletion in the splice donor site of intron 4 (indicated by an asterisk) leads to exon 4 skipping ([Delta]4) (B), cryptic splicing within exon 4 ([Delta]4cryptic ) (C) or cryptic splicing within intron 4 (insTAC ) (D). The deletion of, respectively, 251 or 169 bp in the [Delta]4 and [Delta]4cryptic transcripts results in a frameshift (undulated) and the introduction of a premature stop codon at, respectively, codon 155 for [Delta]4 or134 for [Delta]4cryptic. N-truncated proteins starting at alternative Met codons (AUG) 139 and 210 are indicated. Cryptic splicing within intron 4 results in the in-frame insertion of 3 bp (TAC), leading to an extra Thr codon (ACA) (black box).
Here we report the observation of the intron 4 mutation in four additional unrelated patients with early-onset AD. Also, we provide evidence that the mutation segregates in an autosomal dominant manner and most likely has one common ancestor. Furthermore, we show that the intron 4 mutation gives rise not only to the truncating deletion transcripts ([Delta]4 and [Delta]4cryptic) but also to an in-frame insertion transcript due to alternative splicing using a cryptic splice site in intron 4 (Fig. 1D). In vivo and in vitro analyses of both the deletion and insertion transcripts using cDNA transfection, western blotting, immunoprecipitation and immunocytochemcial analyses indicated that the AD pathophysiology in these patients results from increased A[beta]42 secretion by the insertion transcript but not the deletion transcripts.
RESULTS
Genetic analysis
Initially, we identified the intron 4 mutation in one case (177) with a family history consistent with autosomal dominant inheritance (Fig. 2A) and a second case (142) for which no family history information was available (35; Table 1). Here we have identified the mutation in an additional four unrelated cases with early-onset AD (Table 1). Patient 593 belongs to a brain bank of autopsy-confirmed AD cases (A.Singleton, unpublished data). No data on family history were available for patient 593. Patients 79/95 and TOR122.2 were referred for DNA diagnosis because of their early-onset age and positive family history consistent with autosomal dominant AD (Fig. 2A). Patient 79/95 died at the age of 41 years and had autopsy confirmation of early-onset AD (A.Singleton, unpublished data). No autopsy confirmation is available yet in family TOR122. Patient 160.4.1 belongs to family F105/160 with autosomal dominant AD (Fig. 2A). Disease in family F105 was shown previously to be linked to chromosome 14 (36). Later genealogical studies indicated that families F105 and F160 belonged to the same pedigree (F105/160) (Fig. 2A) and linkage to chromosome 14 was confirmed with marker D14S77, located 50 kb upstream of PSEN1 (37). Autopsy confirmation of AD was also obtained in two further cases, 160.2.02 and 160.2.04 (Fig. 2A), at 42 and 45 years of age, respectively. Mean onset age of AD in family F105/160 was 37 ± 3 years, ranging from 36 to 40 years.
Figure 2. (A) Pedigrees of the autosomal dominant AD families segregating the PSEN1 intron 4 mutation. Underlined numbers refer to individuals that were available for DNA analyses. *, autopsy-confirmed cases; ->, probands. For reasons of confidentiality generation numbers of the at-risk individuals that were analyzed are not provided. (B) Genomic PCR-PvuII analysis of the PSEN1 intron 4 mutation. PSEN1 exon 4 was PCR amplified using flanking primers 4.1 and 4.2 and digested with PvuII. The G deletion in the intron 4 splice donor consensus sequence destroys the PvuII site present in WT DNA, producing an extra PvuII fragment of 371 bp.
Table 1. AD patients with the PSEN1 intron 4 mutation
| Sample | D14S1028 | D14S77 | D14S1004 | D14S1025 | Family history | Age at onset (years) | Age at death (years) |
| 142 | 235-243 | 203-203 | 190-190 | 153-155 | ? | ? | 40 |
| 177 | 231-243 | 203-235 | 190-192 | 151-153 | + | ? | 45 |
| 593 | 227-243 | 203-207 | 190-196 | 149-153 | ? | ? | 42 |
| 79/95 | 235-243 | 203-209 | 190-194 | 151-153 | + | 35 | 41 |
| F160.4.1 | 241-243 | 203-229 | 190-192 | 149-153 | + | 38 | - |
| TOR122.2 | 231-243 | 203-235 | 190-192 | 149-153 | + | 35 | - |
In each patient the intron 4 mutation was observed by PCR amplification of exon 4 followed by PvuII digestion (Fig. 2B). The presence of the mutation was also confirmed by direct PCR sequencing of exon 4 (data not shown). Other mutations in PSEN1 or PSEN2 were excluded by direct PCR sequencing of all coding exons. Further, sequence analysis excluded mutations in exons 16 and 17 of APP. Subsequently, we analyzed all patients and at-risk individuals available in family F105/160 for the intron 4 mutation using PCR-PvuII analysis of exon 4. The mutation was present in three patients and two at-risk individuals and completely segregated with the linked allele (203 bp) of marker D14S77. The current age of the at-risk individuals carrying the mutation is below the mean onset age in family F105/160. The mutation was absent in four at-risk individuals not segregating the 203 bp allele of D14S77. Also, patient 122.1 in family TOR122 carries the intron 4 mutation. In family 79/95, no other individuals were available for DNA analysis.
Interestingly, all mutation carriers were of British ancestry and had very similar ages at onset, in their mid thirties, and died in their mid forties (Table 1). Genotype analysis of polymorphic dinucleotide repeat markers D14S1028, D14S77, D14S1004 and D14S1025 located near the PSEN1 gene (7) showed that all patients shared one common allele at each of these markers (Table 1). Frequency of the shared alleles as calculated in the sample set of the Centre d'Etudes du Polymorphisme Humain (CEPH) (38) are 7 (243 bp allele of D14S1028), 5 (203 bp allele of D14D77), 33 (190 bp allele of D14S1004) and 16% (153 bp allele of D14S1025), respectively.
cDNA analysis
PSEN 1 cDNA was prepared from frozen brain (142, 177, 593 and 79/95) or lymphoblasts (TOR122.2 and F160.3.02) by RT-PCR amplification. We first tested for the presence of the deletion transcripts ([Delta]4 and [Delta]4cryptic) by PCR amplifying a PSEN1 cDNA fragment using primers located in exons 3 and 5. As in our first report, we also observed two shorter PCR fragments of 158 and 76 bp derived from the deleted transcripts c.88-169del ([Delta]4cryptic) and c.88-338del ([Delta]4), respectively (35), as well as the expected PCR fragment of 327 bp corresponding to WT PSEN1. Sequence analysis of these PCR fragments confirmed that they resulted from cryptic splicing within exon 4 ([Delta]4cryptic) or exon 4 skipping ([Delta]4) and that they represented both the VRSQ plus and minus transcripts (data not shown). The short PCR fragments varied largely in relative intensity, suggesting that the concentrations of the deletion transcripts [Delta]4 and [Delta]4cryptic are different in each case. Sequencing of the 327 bp PCR amplification fragment showed a doubling of the sequence pattern near the exon 4-5 junction, suggesting the presence of two sequences (Fig. 3A). Careful analysis of the sequence pattern revealed that apart from the WT PSEN1 transcript, a second transcript was present due to cryptic splicing within intron 4 resulting in the insertion of the 3 bp TAC (c.338-339insTAC, insTAC) (Fig. 3B), which at the protein level corresponds to an in-frame insertion of one Thr (ACA) between amino acids 113 and 114 (T113-114ins) (Fig. 3B). The presence of the insTAC transcript was confirmed by PCR amplification of PSEN1 cDNA using a TAC insertion-specific primer with its 3[prime] end complementary to TAC and a primer located in exon 7. The predicted fragment of 425 bp derived from the insTAC transcript was present in all cases but not in a control (Fig. 3C). To assess the relative quantities of the WT and insTAC PSEN1 cDNA transcripts, we used the GeneScan672 software to detect the PCR amplification products obtained with primers in exons 3 and 5 (Fig. 3D). Measured by the relative peak height, the insTAC transcripts were present at an average concentration of 77 ± 4% of the WT transcript with no significant differences between brain (75 ± 3%, range 72-78%) and lymphoblast (79 ± 5%, range 75-84%) cDNA. In the same analysis, the [Delta]4 and [Delta]4cryptic transcript levels were too low to detect. Also, the relative ratio of 40:60% for the VRSQ plus and minus transcripts was confirmed (7).
Figure 3. cDNA analysis of the PSEN1 intron 4 mutation. (A) Part of the sequencing pattern of the PSEN1 cDNA. Direct sequencing of a 315-330 bp fragment obtained with primers N10F and N11R was performed using sequencing primer 901 located in exon 4 (2). Presence of the insTAC transcripts causes the signal peaks to double at the splice junction between exons 4 and 5. (B) Scheme of the cryptic splicing event leading to the insTAC transcripts. The G deletion in the intron 4 splice donor consensus sequence induces the use of a downstream cryptic splice site resulting in the insertion of an extra TAC nucleotide triplet (underlined). (C) insTAC-specific PCR amplification using primers THR2Fwd and THR2Rev, resulting in a 425 bp PCR fragment when the insTAC transcript is present. (D) Semi-quantitative fluorescent analysis of insTAC and WT transcripts. PCR amplification was performed using primers N10F in exon 3 and N11R in exon 5, resulting in two fragments of 315 and 327 bp, representing the VRSQ plus and minus transcripts of the WT allele, and 318 and 330 bp, for the VRSQ plus and minus transcripts of the insTAC allele. Fragment sizes are indicated in bp.
Sequencing of the complete PSEN1 cDNA in two cases (142 and 177) using overlapping primer sets confirmed the presence of both WT and insTAC transcripts. No other transcripts were detected.
Site-directed mutagenesis
[Delta]4, [Delta]4cryptic and insTAC PSEN1 cDNA constructs were generated by site-directed mutagenesis of the WT PSEN1 cDNA cloned downstream of the cytomegalovirus promoter in pCDNA3. For detection of the PSEN1 proteins we used the N-terminal-specific antisera SB128 and SB129 and the [alpha]PS1loop antiserum (13). SB128 and SB129 are new antisera raised in rabbits against the synthetic peptide LPAPLSYFQNAQMSE, corresponding to amino acids 3-18. Their specificity was demonstrated by pre-adsorbing the antiserum with its antigen on western blots (data not shown). In all experiments we used SB129, but similar results were obtained with SB128.
IVT of WT, [Delta]4 and [Delta]4cryptic cDNA constructs was performed and translation products were analyzed by SDS-PAGE (data not shown). A band of ~7 kDa was observed for [Delta]4cryptic but not for [Delta]4. Using SB129, the ~7 kDa peptide was precipitated, confirming that the ~7 kDa peptide most likely corresponds to the predicted C-truncated peptide starting at Met1. We will refer to this ~7 kDa peptide as PSEN1C-truncated (PSEN1Ctrunc).
cDNA transfections
CHO cells were transiently transfected with WT, [Delta]4, [Delta]4cryptic or insTAC PSEN1 cDNA constructs and analyzed by western blotting with SB129 and [alpha]PS1loop antisera. Immunoblotting with SB129 revealed a band of 46 kDa in WT and insTAC PSEN1 transfectants corresponding to the full-length PSEN1, together with the ~28 kDa NTF (Fig. 4A). In the [Delta]4 or [Delta]4cryptic transfectants the predicted ~7 kDa PSEN1Ctrunc peptide was not observed; only the endogenous ~28 kDa NTF was detected. However, RT-PCR on RNA extracted from the transfected cells indicated that the truncated transcripts were present in the [Delta]4 and [Delta]4cryptic transfectants, suggesting that PSEN1Ctrunc peptides could have been formed. Immunoprecipitation of conditioned medium of transfectants excluded that PSEN1Ctrunc was secreted in the medium. Western blot analysis of the [Delta]4 and [Delta]4cryptic transfectants with [alpha]PS1loop recognized two aberrant bands of ~30 and ~25 kDa, in addition to the ~18 kDa CTF (Fig. 4B). Since the ~30 and ~25 kDa proteins were not recognized by SB129, these proteins are lacking at least part of the N-terminus corresponding to amino acids 3-18, the epitope of SB129. Both bands were also present in the cell lysates of WT and insTAC PSEN1 transfectants, although at significantly lower expression levels compared with the PSEN1 full-length ~46 kDa band. Also, total PSEN1 expression levels in the PSEN1 [Delta]4 and [Delta]4cryptic transfectants were significantly lower than in the WT PSEN1 transfectants. PSEN1 expression levels were also compared between WT, [Delta]4 and [Delta]4cryptic transfectants using immunofluorescent cytochemical analysis using [alpha]PS1loop. Again, reduced expression of PSEN1 was observed in the [Delta]4 and [Delta]4cryptic transfectants versus WT transfectants, confirming the western blot experiments. All of the [Delta]4, [Delta]4cryptic and WT PSEN1 showed perinuclear localization. Furthermore, no immunoreactivity was observed in the [Delta]4 and [Delta]4cryptic transfectants when SB129 was used, confirming the absence of PSEN1Ctrunc. The transient transfection experiments were also performed in HEK293 and Neuro2a cells with similar results on western blot analysis with SB129 and [alpha]PS1loop antisera.
Figure 4. Western blot analysis of PSEN1 cDNA transfectants. (A) An aliquot of 15 µg of cell lysate protein from CHO-K1 cells transfected with empty vector (Mock), human WT PSEN1 cDNA (PSEN1), PSEN1 [Delta]4 cDNA ([Delta]4), PSEN1 [Delta]4cryptic cDNA ([Delta]4cryptic) or PSEN1 insTAC cDNA (insTAC) was separated by SDS-PAGE on a 10-20% polyacrylamide gel and transferred onto nitrocellulose membranes. Membranes were incubated with SB129 antiserum (1:1000) and secondary HRP-conjugated antibody (1:5000) was detected using the super-sensitive chemiluminescent substrate (Pierce). (B) An aliquot of 15 µg of cell lysate protein from CHO-K1 cells transfected with empty vector (Mock), WT PSEN1 cDNA (PSEN1), PSEN1 [Delta]4 cDNA ([Delta]4), PSEN1 [Delta]4cryptic cDNA ([Delta]4cryptic) or PSEN1 insTAC cDNA (insTAC) was separated by SDS-PAGE on a 10% polyacrylamide gel and immunoblotted using [alpha]PS1loop (1:1000) as described in (A). (C) In vitro mutagenesis. Met codons 139, 146 and 210 were changed into Ala codons using site-directed mutagenesis. CHO-K1 cells were transiently transfected with empty vector (Mock), PSEN1 [Delta]4 cDNA ([Delta]4), PSEN1 [Delta]4cryptic cDNA ([Delta]4cryptic), PSEN1 [Delta]4 Met139Ala cDNA (139), PSEN1 [Delta]4 Met146Ala cDNA (146) or PSEN1 [Delta]4cryptic Met210Ala cDNA (210). An aliquot of 15 µg of cell lysate was separated on a 10% polyacrylamide gel and immunoblotted with [alpha]PS1loop. (D) PSEN1 catabolism. HEK-293 WT PSEN1 (PSEN1), PSEN1 [Delta]4 ([Delta]4) and PSEN1 [Delta]4cryptic ([Delta]4cryptic) stable transfectants were treated for 4 h with 50 µM ALLN (+) or with DMSO (-) as a control. An aliquot of 15 µg of cell lysate protein was separated by SDS-PAGE on a 10-20% polyacrylamide gel and immunoblotted with SB129 as described (A). As a result of extended exposure times, endogenous full-length PSEN1 is seen in addition to endogenous NTFs.
Stable HEK-293 cell lines were established for PSEN1 WT, [Delta]4, [Delta]4cryptic and insTAC and analyzed by western blotting using SB129 and [alpha]PS1loop. With [alpha]PS1loop, again the aberrant PSEN1 peptides of ~25 and ~30 kDa were observed in the [Delta]4 and [Delta]4cryptic transfectants in addition to the ~18 kDa CTF (data not shown).
Careful examination of the PSEN1 coding sequence indicated that there are at least three other putative translational initiation sites within favorable Kozak sequences i.e. Met139, Met146 and Met210. Initiation of translation at these sites would produce N-truncated PSEN1 proteins with predicted molecular weights of 36 kDa (Met139 and Met146) and 28.3 kDa (Met210). These sizes are in good agreement with the sizes of the two aberrant PSEN1 proteins recognized in [Delta]4 and [Delta]4cryptic transfectants by [alpha]PS1loop. Also, these aberrant bands are negative for the SB129 N-terminal antiserum. Based on these observations we hypothesized that these aberrant proteins correspond to N-truncated PSEN1 proteins, starting at Met139 or Met146 (~30 kDa) and Met210 (~25 kDa), respectively. To test this hypothesis we used site-directed mutagenesis of the [Delta]4 and [Delta]4cryptic cDNA constructs resulting in the replacement of the Met (ATG) codon by an Ala (GCG) codon. CHO cells were used in transient transfection experiments and cell lysates were analyzed by western blotting with [alpha]PS1loop (Fig. 4C). In the Met139Ala [Delta]4 transfectants the ~30 kDa band decreased in density in favor of the ~25 kDa band, while in Met146Ala [Delta]4 there was no obvious change in the expression levels of the ~25 and ~30 kDa proteins. The ~25 kDa protein is no longer formed in cells transfected with Met210Ala [Delta]4cryptic cDNA. Our mutagenesis data are consistent with the two aberrant proteins of ~30 and ~25 kDa being N-truncated proteins resulting from initiation of translation at respectively Met139 and Met210, but not Met146, in the deletion mutants (Fig. 1B and C). From now on we will refer to these N-truncated PSEN1 proteins as PSEN1Ntrunc.
Next, we examined turnover of the truncated PSEN1 proteins in stable HEK-293 PSEN1 WT, [Delta]4 and [Delta]4cryptic transfectants after treatment with ALLN, a potent inhibitor of the non-lysosomal degradation pathway (Fig. 4D). Western blot analysis of ALLN-treated transfectants with SB129 revealed the ~7 kDa PSEN1Ctrunc in cells expressing [Delta]4cryptic but not in those expressing [Delta]4 (Fig. 4D). These data indicate that PSEN1Ctrunc is formed but rapidly degraded under normal cell culture conditions.
In vivo analysis of PSEN1 expression
Frozen brain tissue of two patients (142 and 177) with the PSEN1 intron 4 mutation (35) was homogenized and analyzed by western blotting and immunoprecipitation with SB129 and [alpha]PS1loop antisera (Fig. 5). No full-length PSEN1 was detected after western blotting; only the ~28 kDa NTF and ~18 kDa CTF fragments derived from conventional proteolysis of PSEN1 were observed. Neither were the PSEN1Ctrunc and PSEN1Ntrunc proteins observed. Immunoprecipitation of 1 mg of brain homogenate did not reveal PSEN1Ctrunc or PSEN1Ntrunc proteins (data not shown).
Figure 5. In vivo detection of PSEN1 proteins in brain homogenates. (A) Aliquots of 100 µg of brain homogenates of two patients (142 and 177) carrying the PSEN1 intron 4 mutation, as well as of a healthy control individual, were separated by SDS-PAGE on a 10-20% polyacrylamide gel and immunoblotted with SB129 as described in Figure 4A. (B) As in (A), but immunoblotted with [alpha]PS1loop.
Lymphoblasts of three patients (160.3.1, 160.3.2 and TOR122.2) as well as of two at-risk individuals from family F105/160 not carrying the intron 4 mutation were analyzed by western blotting with SB129 and [alpha]PS1loop antisera. Again, the PSEN1Ctrunc and PSEN1Ntrunc proteins were not observed in the mutation carriers (data not shown).
A[beta] secretion
A[beta]40 and A[beta]42 concentrations were measured in conditioned medium of HEK-293 cells of stable [Delta]4, [Delta]4cryptic and insTAC transfectants using a prototype version of the INNOTEST [beta]amyloid1-42 HS ELISA test (21,26). For each transfectant we analyzed multiple clones in triplicate. Mean values are given in Table 2. No significant changes in A[beta]40 concentration were observed between WT and mutant transfectants. Also, no significant increase in secreted A[beta]42 concentrations was observed between WT and [Delta]4 (P = 0.5) or between WT and [Delta]4cryptic (P = 0.13). However, a significant increase (~3.4 times) in secreted A[beta]42 was observed between WT and insTAC (P = 0.0013), resulting in a 3.8-fold increase in A[beta]42:A[beta]40 ratio (P = 2 × 10-6). As positive controls we included PSEN1 Y115C and PSEN1 G384A transfectants. The Y115C missense mutation was observed previously in a Dutch family (1066) with mean onset age of AD of 42.5 ± 4.5 years (39,40). The G384A mutation was observed in a Belgian family (AD/B) with mean age of onset 34.7 ± 3.0 years (7,41). A significant increase (~5.4 times) in A[beta]42 concentration and in A[beta]42:A[beta]40 ratio (4.3 times) was also observed for the Y115C transfectants compared with WT (P = 8 × 10-5 and 2 × 10-5, respectively). G384A transfectants analyzed in the same experiment showed a massive increase in A[beta]42 secretion (~20 times), as observed previously (21). The results were reproduced when the experiment was repeated.
Table 2. Secreted A[beta] in stably transfected HEK-293 cells
| A[beta]42 | A[beta]40 | Ratio (%) | n | |
| WT | 9.0 ± 2.8 | 41 ± 12 | 21.76 ± 0.89 | 3 |
| [Delta]4 | 11.0 ± 2.8 | 46 ± 12 | 23.81 ± 0.03 | 2 |
| [Delta]4cryptic | 13.5 ± 0.7 | 54 ± 3 | 25.35 ± 0.27 | 2 |
| T113-114ins | 30.9 ± 6.8 | 38 ± 8 | 81.74 ± 6.97 | 6 |
| Y115C | 49.3 ± 3.2 | 52 ± 1 | 95.16 ± 5.44 | 3 |
| G384A | 187.3 | 76 | 245.04 | 1 |
We also observed an ~2-fold increased A[beta]42:A[beta]40 ratio, although not statistically significant, in brain homogenates from patients 142, 177, 593 and 79/95, carrying the PSEN1 intron 4 mutation, compared with the A[beta]42:A[beta]40 ratio in healthy controls and in sporadic AD patients (data not shown).
DISCUSSION
When we initially reported the intron 4 mutation in two autopsy-confirmed early-onset AD patients (35), we had no absolute proof that this mutation was causally related to the disease, since we could not examine co-segregation of the mutation in the patient families. In this study, we identified the same mutation in an additional four unrelated early-onset AD cases, three of which belonged to autosomal dominant AD pedigrees. All patients had a similar early-onset age of AD (n = 16, mean 37.3 ± 1.4 years, range 35-40 years) and age at death (n = 13, mean 44 2 years, range 40-47 years) with a mean duration of disease of 7 ± 1 years. Also, all patients were of British ancestry, suggesting that the intron 4 mutation is a frequent mutation, at least in Great Britain. Mutation analysis for the intron 4 mutation in a population-based sample of 102 Dutch early-onset AD cases (39) did not detect mutation carriers. Another possibility is that all mutation carriers are distantly related, sharing a common ancestor. This hypothesis was confirmed by analyzing four highly polymorphic dinucleotide repeat markers flanking the PSEN1 gene, showing that all patients shared one common allele at each of the markers (Table 1). Based on the frequencies of the shared alleles and considering each marker independently, the probability of observing this allele combination in six unrelated cases by chance is 10-16.
Based on the observation of the [Delta]4 and [Delta]4cryptic truncated mRNA transcripts in brains of patients carrying the PSEN1 intron 4 mutation, we previously hypothesized that the PSEN1 intron 4 mutation would lead to AD by haploinsufficiency of PSEN1 (35). However, here we report the presence of a third transcript caused by the same mutation. This transcript is due to the insertion of a TAC triplet between exons 4 and 5 of PSEN1 (insTAC) and results in the insertion of a Thr after amino acid 113 (T113-114ins), located in the first hydrophilic loop (HL-I) of the PSEN1 protein. In HL-I, six missense mutations are located in codons 115 (39,42), 117 (43), 120 (44-46) and 123 (47), suggesting that this HL is particularly vulnerable to mutation. Sequencing of the complete PSEN1 cDNA sequence excluded the presence of other PSEN1 transcripts than [Delta]4, [Delta]4cryptic and insTAC. Semi-quantitative fluorescent analysis of the WT and insTAC PSEN1 transcripts indicated that the expression level of the insTAC mutation was ~75% of that of the WT transcript. No significant differences were observed between brain and lymphoblast cDNA. Also, in the same analysis, the [Delta]4 and [Delta]4cryptic deletion transcripts were undetectable.
We generated PSEN1 cDNA constructs corresponding to the [Delta]4, [Delta]4cryptic and insTAC cDNA transcripts by in vitro mutagenesis of full-length PSEN1 cDNA and transfected them into mammalian cells. For the insTAC transcript, we demonstrated the formation of an ~46 kDa protein. This ~46 kDa mutant PSEN1 protein was indistinguishable from the WT PSEN1 by the methods we used. We failed to raise an antibody which specifically and exclusively recognizes the mutant protein using the synthetic peptide CKDGQLTIYTPF. For the [Delta]4 and [Delta]4cryptic truncated transcripts, we showed that the predicted PSEN1 C-truncated peptide of ~7 kDa (PSEN1Ctrunc) is formed, although only in the [Delta]4cryptic transfectants, but is rapidly degraded since it was only detectable after inhibiting the non-lysosomal degradation pathway of the cell. This rapid degradation is not unexpected, because PSEN1Ctrunc is a short peptide containing several hydrophilic or charged residues and is not inserted in the membrane. PSEN1Ctrunc is therefore not likely to be functionally active. In addition to PSEN1Ctrunc, N-truncated PSEN1 peptides of ~30 and ~25 kDa (PSEN1Ntrunc) are formed from both [Delta]4 and [Delta]4cryptic cDNA constructs. By in vitro mutagenesis and cDNA transfection we demonstrated that PSEN1 N-truncation results from initiation of translation at Met139 (TM-II) and to a lesser extent Met210 (TM-IV). For [Delta]4cryptic, Met139 is located downstream of the premature stop codon at codon 134 (Fig. 1), allowing the ribosomes to re-initiate translation at Met139 (or Met210), with the formation of both PSEN1Ctrunc and PSEN1Ntrunc. Translation re-initiation has been documented before for eukaryotic mRNAs containing premature stop codons (48,49). For [Delta]4, however, Met139 is located upstream of the premature stop codon 155 (Fig. 1), implying that PSEN1Ctrunc and PSEN1Ntrunc cannot be translated from the same transcript in this case. Only PSEN1Ntrunc, but no PSEN1Ctrunc, was detected in cells expressing [Delta]4 cDNA, suggesting that translation initiation took place at Met139, skipping Met1 completely. It is known that ribosomal subunits can bypass Met1 and initiate translation at the next favorable Met codon (50-53) and that, in the case of frameshifted transcripts, these alternative Met codons are preferentially used (54). Also, in WT PSEN1-expressing cells minor translation products of ~30 and ~25 kDa were detected that potentially correspond to the N-truncated PSEN1 proteins, but in vitro mutagenesis of WT PSEN1 was not performed. When WT PSEN1 was transiently overexpressed in CHO cells, the N-truncated PSEN1 peptides were also observed, confirming that Met139 and Met210 are suitable translation initiation codons. However, the ~30 and ~25 kDa proteins observed in WT PSEN1-expressing cells are only minor translation products, whereas in [Delta]4 and [Delta]4cryptic transfectants they are the major translation products.
In brain homogenates and lymphoblast lysates of patients with the intron 4 mutation we could not detect the PSEN1Ctrunc or PSEN1Ntrunc proteins. However, we cannot exclude that the truncated PSEN1 proteins are formed at concentrations below the detection limit of the methods used. On the other hand, if PSEN1Ctrunc is formed it is most likely rapidly degraded through the non-lysosomal pathway, as suggested by our in vitro experiments. PSEN1Ntrunc, having, respectively, six and four intact TM domains and potentially being integrated into the membrane, may also be an artifact of the in vitro overexpression system.
Missense PSEN1 mutations are believed to be pathogenic by a gain-of-function in which A[beta]42 secretion is selectively increased (26,29,30). To test whether [Delta]4, [Delta]4cryptic and/or insTAC have a similar effect on A[beta]42 secretion, we measured A[beta]40 and A[beta]42 concentrations in media from [Delta]4, [Delta]4cryptic and insTAC stable transfectants. No significant variations in A[beta]40 were detected between WT and any of the mutant transfectants. Also, we observed no significant increase in A[beta]42 secretion for [Delta]4 and [Delta]4cryptic when compared to WT PSEN1. This result corroborates with the finding that N-truncated PSEN2 proteins, resulting from naturally occurring alternative splicing of PSEN2, do not lead to increased A[beta]42 secretion (55). A significant increase in A[beta]42 secretion was seen, however, for the insTAC transfectants compared with WT PSEN1 transfectants. We also compared the increase in A[beta]42 secretion for the insTAC transfectants with the increase in A[beta]42 concentration caused by a missense mutation Y115C located in HL-I, close to the site of the T113-114ins mutation. The A[beta]42 increase was comparable, although slightly higher for the missense mutation. We also demonstrated an increased A[beta]42:A[beta]40 ratio in the brains of patients carrying the PSEN1 intron 4 mutation. Immunohistochemical analysis of brain tissue of these patients also showed a similar amount and pattern of A[beta]42 deposition in senile plaques as observed in patients carrying PSEN1 missense mutations (A.Singleton, unpublished data). Our in vitro and in vivo data indicate that the truncated proteins formed during overexpression of the [Delta]4 and [Delta]4cryptic transcripts do not have a biological function in A[beta]42 secretion. The PSEN1 T113-114ins formed by the insTAC transcript, however, clearly led to increased A[beta]42 secretion, comparable with PSEN1 missense mutations.
In conclusion, we have provided evidence that the PSEN1 intron 4 mutation is causally related to AD. Furthermore, we have demonstrated the presence of an extra insertion transcript, insTAC, in the brains and/or lymphoblasts of patients carrying this mutation. In vitro analysis of [Delta]4, [Delta]4cryptic and insTAC transfectants suggests that the insTAC transcript is most likely responsible for the AD phenotype in patients carrying the PSEN1 intron 4 mutation. The [Delta]4 and [Delta]4cryptic transcripts, on the other hand, are most probably not important contributors to AD pathophysiology. Therefore, the PSEN1 intron 4 mutation does not lead to AD pathology due to haploinsufficiency of PSEN1, but is pathogenic by a similar mechanism to PSEN1 missense mutations, resulting in increased A[beta]42 secretion.
MATERIALS AND METHODS
Mutation analysis
Genomic DNA was isolated from total blood of patients 79/95, TOR122.2 and 160.4.01 by phenol-chloroform extraction. Frozen brain material was available from patients 142 (cortex), 177 (cortex), 593 (frontal and temporal cortex) and 79/95 (frontal cortex). Lymphoblast cell lines were available from two patients (160.3.01, cell line AC27, and 160.3.02, cell line AC30), six at-risk individuals (cell lines AC53, AC20, AC18, AC22, AC24 and AC0215) and three spouses (cell lines AC0214, AC23 and AC29) from family F105/160. Genomic DNA was isolated from lymphoblast cell lines and autopsied brain material using a Blood Extraction Kit (Qiagen, Hilden, Germany). PCR amplification of a 371 bp fragment containing exon 4 and flanking intronic sequences was performed using primers 4.1 and 4.2 (39). The fragments were digested with 5 U PvuII and separated by electrophoresis for 2 h at 180 V on a 1.5% agarose gel. In the presence of the PSEN1 intron 4 mutation, a PvuII restriction site cutting the fragment into two parts of 308 and 63 bp is abolished (35).
Polymorphism analysis
The STRs D14S1028, D14S77, D14S1004 and D14S1025 were PCR amplified using published primers of which one was fluorescently labeled. The alleles were separated on a 6% polyacrylamide gel containing 8 M urea using an ABI373A automated DNA sequencer (Applied Biosystems, Foster City, CA) and analyzed using the GeneScan 672 software (Applied Biosystems).
cDNA analysis
RNA was isolated from lymphoblast cells and/or homogenized frozen brain tissue using the Superscript Preamplification kit (Gibco BRL, Gaithersburg, MD) and first strand cDNA was synthesized using the random hexamer and oligo(dT) method. Both first strand cDNA preparations were pooled for further analyses using standard PCR amplification reactions.
PSEN1 cDNA fragments were PCR amplified using primers 917 in the 5[prime]-UTR and 892 in exon 5 (2) and then amplified a second time with internal primers N10F in exon 3 and N11R in exon 5 (7) resulting in a WT PSEN1 fragment of 327 bp in the presence of the exon 3 VRSQ motif.
A PCR that specifically amplifies the PSEN1 insTAC mutation was performed using the insertion-specific primer THR2Fwd (5[prime]-CGGAAGGATGGGCAGCTTACA-3[prime]) and primer THR2Rev (5[prime]-AGCCACGCAGTCCATTCAGG-3[prime]). Amplification of the insertion mutation transcript results in a fragment of 425 bp fragment, while the WT and deletion transcripts cannot be amplified.
Sequence analysis
Genomic DNA or cDNA were PCR amplified and sequencing was performed using the Taq Dye Terminator Sequenase II Sequencing Kit (Applied Biosystems). The products were analyzed on an ABI373 automated DNA sequencer (Applied Biosystems).
PSEN1 cDNA constructs
The full-length PSEN1 cDNA (56) was recloned into the BamHI and EcoRI sites of pCDNA3 (Invitrogen, Leek, The Netherlands). Site-directed mutagenesis was performed using the Transformer site-directed mutagenesis system (Clontech, Palo Alto, CA) or the QuikChange site-directed mutagenesis system (Stratagene, La Jolla, CA). Primer 5[prime]-CCACCTGAGCAATACTAATCTATACCCCATTC-3[prime] was used to delete the sequence corresponding to exon 4 and primer 5[prime]-ATGGACGACCCCAGGAATCTATACCCCATTC-3[prime] to remove that part of the exon 4 sequence which is deleted in the [Delta]4cryptic transcript. Primers 5[prime]-cccggaaggatgggcagcttacaatctataccccattcac-3[prime] and 5[prime]-gaatgggtatagattgtaagctgcccatccttccggg-3[prime] were used to insert ACA in the insTAC transcript.
Sequencing of the [Delta]4 and [Delta]4cryptic cDNA constructs confirmed that their sequence was in full agreement with that of the mutant PSEN1 cDNAs, present in the brains of the two AD patients with the intron 4 mutation (35). Primers 5[prime]-GCTGCCATCGCGATCAGTGTC-3[prime] and 5[prime]-GTGTGGTGGGAGCGATTTCCATTC-3[prime] were used to replace the Met codons by Ala codons at positions 139 and 210, respectively, in the truncated constructs. A complementary primer pair, harboring the mutation at codon 146 (mut146-S, 5[prime]-CATTGTTGTCGCGACTATCCTCC-3[prime]; mut146-AS, 5[prime]-GGAGGATAGTCGCGACAACAATG-3[prime]) as well as a 5[prime] PSEN1-specific primer 201-S (5[prime]-TCAAGAGGCTTTGTTTTCTG-3[prime]) and a 3[prime] PSEN1-specific primer 1742-AS (5[prime]-CGGGAATCTTGACTTTGTTA-3[prime]) were used to introduce the Met146Ala mutation in the truncated constructs in a two-stage PCR reaction: in the first stage, 5 ng [Delta]4 cDNA plasmid were used in two standard PCR reactions of 30 cycles (30 s at 94°C, 60 s at 58°C, 30 s at 72°C) with 25 pmol of, respectively, primer pair 201-S and mut146-AS or mut146-S and 1742-AS in 10 mM KCl, 10 mM (NH4)2SO4, 20 mM Tris-HCl pH 8.8, 2 mM MgSO4, 0.1% Triton X-100, 200 µM dNTPs, 0.6 U Vent polymerase (New England Biolabs, Beverly, MA). After agarose electrophoresis, DNA bands were excised and the DNA was eluted over glass wool. Aliquots of 5 µl of each eluate were combined and used as template in a second PCR reaction with 10 pmol 5[prime]-phosphorylated primer 201-S and 5[prime]-phosphorylated primer 1742-AS in 10 mM KCl, 10 mM (NH4)2SO4, 20 mM Tris-HCl pH 8.8, 2 mM MgSO4, 0.1% Triton X-100, 200 µM dNTPs, 0.6 U Vent polymerase (New England Biolabs). The purified PCR product was ligated overnight with T4 DNA ligase (Gibco BRL, Bethesda, MD) in EcoRV dephosphorylated pCDNA3 vector.
In vitro transcription/translation
An aliquot of 1 µg recombinant plasmid DNA was transcribed from the T7 promoter and translated in the presence of 40 µCi [35S]methionine using the TNT T7 coupled reticulocyte lysate system (Promega, Madison, WI). Samples of 1/50 of the final reaction product were separated on a 10-20% Tris-glycine gel (Novex, San Diego, CA), fixed in a 65:25:10 mixture of isopropanol, water and acetic acid and dried under vacuum. Autoradiography was performed overnight.
Transfection
Chinese hamster ovary K1 (CHO-K1) cells and Neuro 2a mouse neuroblastoma cells were transfected with pCDNA3 vector (Mock) or with pCDNA3 containing PSEN1 cDNAs using lipofectamine (Gibco BRL) according to the manufacturer's procedures. For transfection of human embryonic kidney (HEK-293) cells, DOTAP (Boehringer Mannheim, Mannheim, Germany) was used. At 24-48 h post-transfection, cells were analyzed for protein expression either by immunofluorescence microscopy, immunoblotting or immunoprecipitation.
HEK-293 cells, stably expressing PSEN1 WT, [Delta]4, [Delta]4cryptic, insTAC, Y115C or G384A were selected for integration of the recombinant plasmid in the cell genome by resistance to 800 µg/ml G418.
cDNA synthesis
At 24 h after transient transfection, RNA was isolated using RNAzol (Campro Scientific, Veenendaal, The Netherlands). Samples of 1 µg RNA were incubated with 2 U RQ1 RNase-free DNase (Promega) in 40 mM Tris-HCl pH 8.4, 100 mM KCl, 5 mM MgCl2 for 10 min at 37°C. First strand cDNA was made with random hexamer primers using the SuperScript preamplification system (Gibco BRL). Aliquots of 1.5 µl of the final reaction product were used as template in a standard PCR reaction of 30 cycles (60 s at 94°C, 90 s at 58°C, 90 s at 72°C) in 20 mM Tris-HCl pH 8.4, 50 mM KCl, 1 mM MgCl2, 25 pmol of the primer pair N8F (7) and 892 (2), 200 µM dNTPs, 0.2 U Taq polymerase (Gibco BRL). The PCR products were analyzed on a 2% agarose gel.
Immunoblot analysis
Transfected cells were washed in phosphate-buffered saline (PBS) and lysed in SDS lysis buffer [2% SDS in 150 mM NaCl, 50 mM Tris-HCl pH 8.0, containing 2× complete protease inhibitors (Boehringer Mannheim)]. Human lymphoblasts, maintained in culture in RPMI 1640 medium (Gibco BRL) supplemented with 15% fetal bovine serum and antibiotics, were harvested, washed and lysed in RIPA buffer (50 mM Tris pH 8.0, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS with complete protease inhibitors). A 10% brain homogenate was made in TNE buffer (50 mM Tris pH 8.0, 150 mM NaCl, 5 mM EDTA) containing 4% SDS and complete protease inhibitors (Boehringer Mannheim). Aliquots of 10-40 µg cell lysate or 100 µg brain homogenate were separated on a 10% Tris-tricine, a 12% Tris-glycine or a 10-20% gradient Tris-tricine gel (Novex, San Diego, CA), electroblotted onto Hybond C membranes (Amersham, Aylesbury, UK) and incubated with SB129 (1:750), N-terminal PSEN1 (1:1000) or [alpha]PS1loop (1:1000), followed by incubation with horseradish peroxidase (HRP)-coupled anti-rabbit Fab2 fragment antibody (1:5000) or HRP-coupled protein A (1:500), respectively. Antibody binding was visualized using the Super Signal Chemiluminescent substrate (Pierce, Rockford, IL) or ECL (Amersham).
Immunoprecipitation
An aliquot of 50 µl of IVT reaction product was diluted in 500 µl RIPA buffer. To this were added 3 µl SB129 antiserum and 20 µl Sepharose A beads (Boehringer Mannheim) with rotation overnight at 4°C. Sepharose A beads were spun off for 2 min at 720 g and 4°C, and washed three times in RIPA buffer. The beads were well mixed with 20 µl of 2× sample buffer (0.9 M Tris-HCl pH 8.45, 20% glycerol, 8% SDS, 0.015% Coomassie blue, 0.05% Phenol red, 10% [beta]-mercaptoethanol) and incubated for 30 min at 37°C before loading onto a SDS-PAGE gel.
Brain homogenates (10%) were made in RIPA buffer. To 1 mg of brain homogenate were added 3 µl SB129 or [alpha]PS1loop antiserum and 20 µl Sepharose A beads in the presence of an extra 500 µl RIPA buffer and immunoprecipitation was performed as described above.
A sample of 1 ml of conditioned medium of transfected cells, expressing PSEN1 [Delta]4 or PSEN1 [Delta]4cryptic, was made into 1× RIPA buffer and immunoprecipitation with SB129 was performed as described above.
Immunofluorescence microscopy
Cells were grown and transfected on coverslips and fixed for 2 min in ice-cold methanol, blocked for 1 h in TBS, 1% bovine serum albumin, 5% normal swine serum and incubated for 1 h at 37°C with primary antiserum (SB129 1:200; [alpha]PS1loop 1:1000). Cells were washed three times with TBS + 1% BSA and incubated with secondary FITC or Texas Red conjugated antibody (1:200; Molecular Probes, Eugene, OR) for 1 h at room temperature. Cells were washed three times with distilled water and incubated with DAPI (1:5000) for 5 min at room temperature. A Zeiss Axioskop fluorescence microscope was used for visualization of the fluorescent signal.
A[beta] measurement
Stable cell lines were seeded at 2 × 105 cells/10 cm2 well. Medium was conditioned for 48 h. A[beta]42 concentrations in the conditioned medium were measured by sandwich-type enzyme-linked immunosorbent assay (ELISA), using a prototype version of the INNOTEST [beta]-amyloid1-42 HS ELISA (21). A[beta]40 was measured by sandwich-type ELISA, using one polyclonal antibody, R209, as capturing antibody and biotinylated 3D6 (57) as detector antibody. No cross-reactivity was observed with A[beta]1-42 or shorter peptides. HPLC purified A[beta]40 (Bachem, Heidelberg, Germany) was used as standard. The assay was performed as described previously (58), with some modifications. In brief, immunoplates are coated for 2 h at room temperature with a goat anti-rabbit polyclonal antibody (Jackson Laboratories, Bar Harbor, ME). After a wash step, plates were blocked for 1 h at 25°C with PBS, 0.1% casein. Thereafter, plates were incubated for an additional 1 h with a 1/500 dilution of rabbit serum R209. Samples of 100 µl or standards were incubated for 2 h. After decanting and washing, peptide bound to the antibody-coated plate was detected using biotinylated 3D6. After several wash steps, the amount of bound antibody was verified by adding 100 µl HRP-streptavidin (RDI, Flanders, NY). Incubation was continued for 30 min at 25°C. Then, 100 µl of 3,5,3[prime],5[prime]-tetramethylbenzidine in substrate buffer was added as peroxidase substrate. The reaction was stopped after 30 min with 50 µl of 0.9 N H2SO4.
A two-tailed unpaired t-test was used to compare the mean level of A[beta] produced by the WT and mutant transfectants.
ACKNOWLEDGEMENTS
We thank S.S. Sisodia and G. Thinakaran for providing us with the [alpha]PS1loop antiserum and P. Mehta for the R209 A[beta]40 antiserum. We thank L. Hendriks and Y. Liang for their contribution to this work. This work was supported by the Fund for Scientific Research-Flanders (Belgium) (FWO-F), the Flemish Biotechnology Program COT-04, the Interuniversity Attraction Poles (IUAP P4/17), the International Alzheimer's Research Foundation (IARF), the EC Biotech Program (BIO2-CT96-0743), the Focused Giving Program of Johnson & Johnson, Addenbrooke's NHS Trust, the Medical Research Council, the States Education Council Guernsey and the Mayo/USF Program Project grant on the presenilins (J.H.). C.D.J. is a research assistant and M.C. a post-doctoral fellow of the FWO-F. D.C.R. is a Glaxo Wellcome Research Fellow.
REFERENCES
*These authors contributed equally to this work
+To whom correspondence should be addressed. Tel: +32 3 820 26 01; Fax: +32 3 820 25 41; Email: cvbroeck{at}uia.ua.ac.be
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Kralovicova, M. B. Christensen, and I. Vorechovsky Biased exon/intron distribution of cryptic and de novo 3' splice sites Nucleic Acids Res., September 1, 2005; 33(15): 4882 - 4898. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kondo, N. Yamamoto, T. Murakami, M. Okumura, A. Mayeda, and K. Imaizumi Tra2{beta}, SF2/ASF and SRp30c modulate the function of an exonic splicing enhancer in exon 10 of tau pre-mRNA Genes Cells, February 1, 2004; 9(2): 121 - 130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Heckmann, W.-C. Low, C. de Villiers, S. Rutherfoord, A. Vorster, H. Rao, C. M. Morris, R. S. Ramesar, and R. N. Kalaria Novel presenilin 1 mutation with profound neurofibrillary pathology in an indigenous Southern African family with early-onset Alzheimer's disease Brain, January 1, 2004; 127(1): 133 - 142. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Nyabi, M. Bentahir, K. Horre, A. Herreman, N. Gottardi-Littell, C. Van Broeckhoven, P. Merchiers, K. Spittaels, W. Annaert, and B. De Strooper Presenilins Mutated at Asp-257 or Asp-385 Restore Pen-2 Expression and Nicastrin Glycosylation but Remain Catalytically Inactive in the Absence of Wild Type Presenilin J. Biol. Chem., October 31, 2003; 278(44): 43430 - 43436. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. Eckman, M. Watson, L. Marlow, K. Sambamurti, and C. B. Eckman Alzheimer's Disease beta -Amyloid Peptide Is Increased in Mice Deficient in Endothelin-converting Enzyme J. Biol. Chem., January 17, 2003; 278(4): 2081 - 2084. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Dermaut, S. Kumar-Singh, C. De Jonghe, M. Cruts, A. Lofgren, U. Lubke, P. Cras, R. Dom, P. P. De Deyn, J. J. Martin, et al. Cerebral amyloid angiopathy is a pathogenic lesion in Alzheimer's disease due to a novel presenilin 1 mutation Brain, December 1, 2001; 124(12): 2383 - 2392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Athan, J. Williamson, A. Ciappa, V. Santana, S. N. Romas, J. H. Lee, H. Rondon, R. A. Lantigua, M. Medrano, M. Torres, et al. A Founder Mutation in Presenilin 1 Causing Early-Onset Alzheimer Disease in Unrelated Caribbean Hispanic Families JAMA, November 14, 2001; 286(18): 2257 - 2263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Janssen, P. L. Lantos, N. C. Fox, R. J. Harvey, J. Beck, A. Dickinson, T. A. Campbell, J. Collinge, D. P. Hanger, L. Cipolotti, et al. Autopsy-Confirmed Familial Early-Onset Alzheimer Disease Caused by the L153V Presenilin 1 Mutation Arch Neurol, June 1, 2001; 58(6): 953 - 958. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Singleton, R. Hall, C. G. Ballard, R. H. Perry, J. H. Xuereb, D. C. Rubinsztein, C. Tysoe, P. Matthews, B. Cordell, S. Kumar-Singh, et al. Pathology of early-onset Alzheimer's disease cases bearing the Thr113-114ins presenilin-1 mutation Brain, December 1, 2000; 123(12): 2467 - 2474. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kumar-Singh, C. De Jonghe, M. Cruts, R. Kleinert, R. Wang, M. Mercken, B. De Strooper, H. Vanderstichele, A. Lofgren, I. Vanderhoeven, et al. Nonfibrillar diffuse amyloid deposition due to a {gamma}42-secretase site mutation points to an essential role for N-truncated A{beta}42 in Alzheimer's disease Hum. Mol. Genet., November 1, 2000; 9(18): 2589 - 2598. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Janssen, M. Hall, N. C. Fox, R. J. Harvey, J. Beck, A. Dickinson, T. Campbell, J. Collinge, P. L. Lantos, L. Cipolotti, et al. Alzheimer's disease due to an intronic presenilin-1 (PSEN1 intron 4) mutation: A clinicopathological study Brain, May 1, 2000; 123(5): 894 - 907. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sato, K. Imaizumi, T. Manabe, M. Taniguchi, J. Hitomi, T. Katayama, T. Yoneda, T. Morihara, Y. Yasuda, T. Takagi, et al. Increased Production of beta -Amyloid and Vulnerability to Endoplasmic Reticulum Stress by an Aberrant Spliced Form of Presenilin 2 J. Biol. Chem., January 12, 2001; 276(3): 2108 - 2114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Steiner, T. Revesz, M. Neumann, H. Romig, M. G. Grim, B. Pesold, H. A. Kretzschmar, J. Hardy, J. L. Holton, R. Baumeister, et al. A Pathogenic Presenilin-1 Deletion Causes Abberrant Abeta 42 Production in the Absence of Congophilic Amyloid Plaques J. Biol. Chem., March 2, 2001; 276(10): 7233 - 7239. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||